# AISIX (api7) benchmark measurements

Operated by the Busbar team; Busbar is also an entrant. Gateway overhead against a local mock is not end-to-end model latency.  [Methodology and limitations](https://onthebench.ai/gateways/method) .

Rust · recorded build: ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a. Measured: 2026-11-10T17:39:23Z.

[JSON](https://onthebench.ai/api/v1/gateways/aisix)  ·  [Markdown](https://onthebench.ai/gateways/aisix.md)  ·  [CSV](https://onthebench.ai/gateways/data.csv?gateway=aisix&compact=1)  ·  [Project source](https://github.com/api7/aisix)  ·  [Interactive dashboard](https://onthebench.ai/gateways)

Each row below names its protocol pair. A representative dashboard cell may differ between gateways; compare an identical pair before interpreting a ranking. Unavailable metrics retain their reason; below-resolution readings are not exact zeroes.

## Published measurements

| Ingress → upstream | Metric | Value | State / reason | p99 bound (ms) |
| --- | --- | --- | --- | --- |
| openai → anthropic | added_latency_p50_us | 162 | valid; gateway_c1_samples=32039; direct_c1_samples=261797 | — |
| openai → anthropic | added_latency_p99_us | 190 | valid; gateway_c1_samples=32039; direct_c1_samples=261797 | — |
| openai → anthropic | gateway_c1_p99_us | 218 | valid; gateway_c1_samples=32039 | — |
| openai → anthropic | direct_c1_p99_us | 28 | valid; direct_c1_samples=261797 | — |
| openai → anthropic | cpu_us_per_request | 158.5811513058638 | valid | — |
| openai → anthropic | rps_per_cpu_second | 6305.919661733615 | valid | — |
| openai → anthropic | cost_window_conc | 8 | valid | — |
| openai → anthropic | cost_core_utilisation | 0.7859282595133944 | valid | — |
| openai → anthropic | cost_window_ok | 119308 | valid | — |
| openai → anthropic | cost_window_rps | 19882 | valid | — |
| openai → anthropic | cost_threads | 16 | valid | — |
| openai → anthropic | cost_nonvol_ctxt_per_request | 0.000008381667616588997 | valid | — |
| openai → anthropic | cost_majflt | 0 | valid | — |
| openai → anthropic | added_ttft_p50_us | 294 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → anthropic | added_ttft_p99_us | 500 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → anthropic | added_gap_p50_us | 9 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → anthropic | added_gap_p99_us | 19 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → anthropic | streams_sustained_fps | 34767.76115030459 | valid | — |
| openai → anthropic | streams_sustained | 775 | valid | — |
| openai → anthropic | idle_rss_mib | 34.359375 | valid | — |
| openai → anthropic | steady_state_rss_mib | 67.6328125 | valid | — |
| openai → anthropic | recovered_rss_mib | 54.01953125 | valid | — |
| openai → anthropic | peak_rss_mib | 69.20703125 | valid | — |
| openai → anthropic | peak_rss_hwm_mib | 68.87890625 | valid | — |
| openai → anthropic | time_to_plateau_s | not_measured | not_measured | — |
| openai → anthropic | growth_rate_mib_per_min | 0.10804699761897836 | valid | — |
| openai → anthropic | load_s | 300 | valid | — |
| openai → anthropic | shape | 0 | valid | — |
| openai → anthropic | idle_shape | not_measured | not_measured | — |
| openai → anthropic | idle_static | 1 | valid | — |
| openai → anthropic | idle_growth_rate_mib_per_min | -0.6871345013915122 | valid | — |
| openai → anthropic | throughput_rps | 26114 req/s | concurrency 16; samples 156705 | 1 |
| openai → anthropic | throughput_rps | 28484 req/s | concurrency 64; samples 170915 | 5 |
| openai → anthropic | throughput_rps | 29206 req/s | concurrency 128; samples 175266 | 10 |
| openai → anthropic | throughput_rps | 29206 req/s | concurrency 128; samples 175266 | 50 |
| openai → anthropic | throughput_rps | 29206 req/s | concurrency 128; samples 175266 | 100 |
| openai → anthropic | throughput_rps | 29206 req/s | concurrency 128; samples 175266 | unbounded |
| bedrock → anthropic | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| gemini → anthropic | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → anthropic | added_latency_p50_us | 150 | valid; gateway_c1_samples=34173; direct_c1_samples=257073 | — |
| anthropic → anthropic | added_latency_p99_us | 175 | valid; gateway_c1_samples=34173; direct_c1_samples=257073 | — |
| anthropic → anthropic | gateway_c1_p99_us | 204 | valid; gateway_c1_samples=34173 | — |
| anthropic → anthropic | direct_c1_p99_us | 29 | valid; direct_c1_samples=257073 | — |
| anthropic → anthropic | cpu_us_per_request | 150.0564829923434 | valid | — |
| anthropic → anthropic | rps_per_cpu_second | 6664.157256378085 | valid | — |
| anthropic → anthropic | cost_window_conc | 8 | valid | — |
| anthropic → anthropic | cost_core_utilisation | 0.9930564120096029 | valid | — |
| anthropic → anthropic | cost_window_ok | 159340 | valid | — |
| anthropic → anthropic | cost_window_rps | 26553 | valid | — |
| anthropic → anthropic | cost_threads | 16 | valid | — |
| anthropic → anthropic | cost_nonvol_ctxt_per_request | 0 | valid | — |
| anthropic → anthropic | cost_majflt | 0 | valid | — |
| anthropic → anthropic | added_ttft_p50_us | 271 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → anthropic | added_ttft_p99_us | 444 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → anthropic | added_gap_p50_us | 3 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| anthropic → anthropic | added_gap_p99_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| anthropic → anthropic | streams_sustained_fps | 32295.15318055963 | valid | — |
| anthropic → anthropic | streams_sustained | 677 | valid | — |
| anthropic → anthropic | idle_rss_mib | 34.68359375 | valid | — |
| anthropic → anthropic | steady_state_rss_mib | 68.68359375 | valid | — |
| anthropic → anthropic | recovered_rss_mib | 52.7578125 | valid | — |
| anthropic → anthropic | peak_rss_mib | 70.72265625 | valid | — |
| anthropic → anthropic | peak_rss_hwm_mib | 70.5 | valid | — |
| anthropic → anthropic | time_to_plateau_s | not_measured | not_measured | — |
| anthropic → anthropic | growth_rate_mib_per_min | 0.3248954130092891 | valid | — |
| anthropic → anthropic | load_s | 300 | valid | — |
| anthropic → anthropic | shape | 0 | valid | — |
| anthropic → anthropic | idle_shape | not_measured | not_measured | — |
| anthropic → anthropic | idle_static | 1 | valid | — |
| anthropic → anthropic | idle_growth_rate_mib_per_min | -0.6867965671233695 | valid | — |
| anthropic → anthropic | throughput_rps | 28154 req/s | concurrency 16; samples 168946 | 1 |
| anthropic → anthropic | throughput_rps | 30535 req/s | concurrency 64; samples 183220 | 5 |
| anthropic → anthropic | throughput_rps | 31272 req/s | concurrency 128; samples 187661 | 10 |
| anthropic → anthropic | throughput_rps | 31272 req/s | concurrency 128; samples 187661 | 50 |
| anthropic → anthropic | throughput_rps | 31272 req/s | concurrency 128; samples 187661 | 100 |
| anthropic → anthropic | throughput_rps | 31272 req/s | concurrency 128; samples 187661 | unbounded |
| openai-responses → anthropic | added_latency_p50_us | 182 | valid; gateway_c1_samples=28838; direct_c1_samples=259566 | — |
| openai-responses → anthropic | added_latency_p99_us | 215 | valid; gateway_c1_samples=28838; direct_c1_samples=259566 | — |
| openai-responses → anthropic | gateway_c1_p99_us | 244 | valid; gateway_c1_samples=28838 | — |
| openai-responses → anthropic | direct_c1_p99_us | 29 | valid; direct_c1_samples=259566 | — |
| openai-responses → anthropic | cpu_us_per_request | 181.1190625143049 | valid | — |
| openai-responses → anthropic | rps_per_cpu_second | 5521.2299915754 | valid | — |
| openai-responses → anthropic | cost_window_conc | 8 | valid | — |
| openai-responses → anthropic | cost_core_utilisation | 0.9860605945290514 | valid | — |
| openai-responses → anthropic | cost_window_ok | 131074 | valid | — |
| openai-responses → anthropic | cost_window_rps | 21842 | valid | — |
| openai-responses → anthropic | cost_threads | 16 | valid | — |
| openai-responses → anthropic | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai-responses → anthropic | cost_majflt | 0 | valid | — |
| openai-responses → anthropic | added_ttft_p50_us | untestable | untestable | — |
| openai-responses → anthropic | added_ttft_p99_us | untestable | untestable | — |
| openai-responses → anthropic | added_gap_p50_us | untestable | untestable | — |
| openai-responses → anthropic | added_gap_p99_us | untestable | untestable | — |
| openai-responses → anthropic | streams_sustained_fps | untestable | untestable | — |
| openai-responses → anthropic | streams_sustained | untestable | untestable | — |
| openai-responses → anthropic | idle_rss_mib | 34.4609375 | valid | — |
| openai-responses → anthropic | steady_state_rss_mib | 65.390625 | valid | — |
| openai-responses → anthropic | recovered_rss_mib | 52.29296875 | valid | — |
| openai-responses → anthropic | peak_rss_mib | 67.2421875 | valid | — |
| openai-responses → anthropic | peak_rss_hwm_mib | 67.08203125 | valid | — |
| openai-responses → anthropic | time_to_plateau_s | not_measured | not_measured | — |
| openai-responses → anthropic | growth_rate_mib_per_min | -0.11263145814918507 | valid | — |
| openai-responses → anthropic | load_s | 300 | valid | — |
| openai-responses → anthropic | shape | 0 | valid | — |
| openai-responses → anthropic | idle_shape | -1 | valid | — |
| openai-responses → anthropic | idle_static | 0 | valid | — |
| openai-responses → anthropic | idle_growth_rate_mib_per_min | -0.8321574586867803 | valid | — |
| openai-responses → anthropic | throughput_rps | 22218 req/s | concurrency 8; samples 133328 | 1 |
| openai-responses → anthropic | throughput_rps | 25257 req/s | concurrency 64; samples 151548 | 5 |
| openai-responses → anthropic | throughput_rps | 25869 req/s | concurrency 128; samples 155239 | 10 |
| openai-responses → anthropic | throughput_rps | 25869 req/s | concurrency 128; samples 155239 | 50 |
| openai-responses → anthropic | throughput_rps | 25869 req/s | concurrency 128; samples 155239 | 100 |
| openai-responses → anthropic | throughput_rps | 25869 req/s | concurrency 128; samples 155239 | unbounded |
| cohere → anthropic | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| cohere → openai | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| bedrock → openai | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → openai | added_latency_p50_us | 152 | valid; gateway_c1_samples=34038; direct_c1_samples=258158 | — |
| anthropic → openai | added_latency_p99_us | 178 | valid; gateway_c1_samples=34038; direct_c1_samples=258158 | — |
| anthropic → openai | gateway_c1_p99_us | 207 | valid; gateway_c1_samples=34038 | — |
| anthropic → openai | direct_c1_p99_us | 29 | valid; direct_c1_samples=258158 | — |
| anthropic → openai | cpu_us_per_request | 152.09601844195544 | valid | — |
| anthropic → openai | rps_per_cpu_second | 6574.794069192751 | valid | — |
| anthropic → openai | cost_window_conc | 8 | valid | — |
| anthropic → openai | cost_core_utilisation | 0.7563901876709777 | valid | — |
| anthropic → openai | cost_window_ok | 119727 | valid | — |
| anthropic → openai | cost_window_rps | 19952 | valid | — |
| anthropic → openai | cost_threads | 16 | valid | — |
| anthropic → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| anthropic → openai | cost_majflt | 0 | valid | — |
| anthropic → openai | added_ttft_p50_us | 275 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → openai | added_ttft_p99_us | 461 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → openai | added_gap_p50_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| anthropic → openai | added_gap_p99_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| anthropic → openai | streams_sustained_fps | 31294.445560658252 | valid | — |
| anthropic → openai | streams_sustained | 658 | valid | — |
| anthropic → openai | idle_rss_mib | 34.21875 | valid | — |
| anthropic → openai | steady_state_rss_mib | 67.296875 | valid | — |
| anthropic → openai | recovered_rss_mib | 51.80859375 | valid | — |
| anthropic → openai | peak_rss_mib | 68.921875 | valid | — |
| anthropic → openai | peak_rss_hwm_mib | 68.55859375 | valid | — |
| anthropic → openai | time_to_plateau_s | not_measured | not_measured | — |
| anthropic → openai | growth_rate_mib_per_min | -0.5069564534534243 | valid | — |
| anthropic → openai | load_s | 300 | valid | — |
| anthropic → openai | shape | 0 | valid | — |
| anthropic → openai | idle_shape | -1 | valid | — |
| anthropic → openai | idle_static | 0 | valid | — |
| anthropic → openai | idle_growth_rate_mib_per_min | -0.9203045680697568 | valid | — |
| anthropic → openai | throughput_rps | 27958 req/s | concurrency 16; samples 167769 | 1 |
| anthropic → openai | throughput_rps | 30196 req/s | concurrency 64; samples 181184 | 5 |
| anthropic → openai | throughput_rps | 31041 req/s | concurrency 128; samples 186278 | 10 |
| anthropic → openai | throughput_rps | 31041 req/s | concurrency 128; samples 186278 | 50 |
| anthropic → openai | throughput_rps | 31041 req/s | concurrency 128; samples 186278 | 100 |
| anthropic → openai | throughput_rps | 31041 req/s | concurrency 128; samples 186278 | unbounded |
| gemini → openai | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai-responses → openai | added_latency_p50_us | 169 | valid; gateway_c1_samples=30866; direct_c1_samples=260253 | — |
| openai-responses → openai | added_latency_p99_us | 199 | valid; gateway_c1_samples=30866; direct_c1_samples=260253 | — |
| openai-responses → openai | gateway_c1_p99_us | 228 | valid; gateway_c1_samples=30866 | — |
| openai-responses → openai | direct_c1_p99_us | 29 | valid; direct_c1_samples=260253 | — |
| openai-responses → openai | cpu_us_per_request | 162.11809131479575 | valid | — |
| openai-responses → openai | rps_per_cpu_second | 6168.343038644785 | valid | — |
| openai-responses → openai | cost_window_conc | 8 | valid | — |
| openai-responses → openai | cost_core_utilisation | 0.7846584893414216 | valid | — |
| openai-responses → openai | cost_window_ok | 116520 | valid | — |
| openai-responses → openai | cost_window_rps | 19417 | valid | — |
| openai-responses → openai | cost_threads | 16 | valid | — |
| openai-responses → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai-responses → openai | cost_majflt | 0 | valid | — |
| openai-responses → openai | added_ttft_p50_us | untestable | untestable | — |
| openai-responses → openai | added_ttft_p99_us | untestable | untestable | — |
| openai-responses → openai | added_gap_p50_us | untestable | untestable | — |
| openai-responses → openai | added_gap_p99_us | untestable | untestable | — |
| openai-responses → openai | streams_sustained_fps | untestable | untestable | — |
| openai-responses → openai | streams_sustained | untestable | untestable | — |
| openai-responses → openai | idle_rss_mib | 34.46875 | valid | — |
| openai-responses → openai | steady_state_rss_mib | 66.85546875 | valid | — |
| openai-responses → openai | recovered_rss_mib | 52.69921875 | valid | — |
| openai-responses → openai | peak_rss_mib | 68.234375 | valid | — |
| openai-responses → openai | peak_rss_hwm_mib | 68.01171875 | valid | — |
| openai-responses → openai | time_to_plateau_s | not_measured | not_measured | — |
| openai-responses → openai | growth_rate_mib_per_min | 1.2399358321787064 | valid | — |
| openai-responses → openai | load_s | 300 | valid | — |
| openai-responses → openai | shape | 0 | valid | — |
| openai-responses → openai | idle_shape | -1 | valid | — |
| openai-responses → openai | idle_static | 0 | valid | — |
| openai-responses → openai | idle_growth_rate_mib_per_min | -0.8168488447601818 | valid | — |
| openai-responses → openai | throughput_rps | 25449 req/s | concurrency 16; samples 152712 | 1 |
| openai-responses → openai | throughput_rps | 27472 req/s | concurrency 64; samples 164835 | 5 |
| openai-responses → openai | throughput_rps | 28458 req/s | concurrency 128; samples 170775 | 10 |
| openai-responses → openai | throughput_rps | 28458 req/s | concurrency 128; samples 170775 | 50 |
| openai-responses → openai | throughput_rps | 28458 req/s | concurrency 128; samples 170775 | 100 |
| openai-responses → openai | throughput_rps | 28458 req/s | concurrency 128; samples 170775 | unbounded |
| openai → openai | added_latency_p50_us | 146 | valid; gateway_c1_samples=35048; direct_c1_samples=262627 | — |
| openai → openai | added_latency_p99_us | 174 | valid; gateway_c1_samples=35048; direct_c1_samples=262627 | — |
| openai → openai | gateway_c1_p99_us | 202 | valid; gateway_c1_samples=35048 | — |
| openai → openai | direct_c1_p99_us | 28 | valid; direct_c1_samples=262627 | — |
| openai → openai | cpu_us_per_request | 152.64101169979259 | valid | — |
| openai → openai | rps_per_cpu_second | 6551.319261213721 | valid | — |
| openai → openai | cost_window_conc | 8 | valid | — |
| openai → openai | cost_core_utilisation | 0.9445269209542334 | valid | — |
| openai → openai | cost_window_ok | 148977 | valid | — |
| openai → openai | cost_window_rps | 24826 | valid | — |
| openai → openai | cost_threads | 16 | valid | — |
| openai → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → openai | cost_majflt | 0 | valid | — |
| openai → openai | added_ttft_p50_us | 277 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_ttft_p99_us | 366 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_gap_p50_us | 2 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | added_gap_p99_us | below_resolution | below_resolution; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → openai | streams_sustained_fps | 30259.205544499786 | valid | — |
| openai → openai | streams_sustained | 645 | valid | — |
| openai → openai | idle_rss_mib | 34.2890625 | valid | — |
| openai → openai | steady_state_rss_mib | 68.62890625 | valid | — |
| openai → openai | recovered_rss_mib | 52.296875 | valid | — |
| openai → openai | peak_rss_mib | 70.42578125 | valid | — |
| openai → openai | peak_rss_hwm_mib | 70.28515625 | valid | — |
| openai → openai | time_to_plateau_s | not_measured | not_measured | — |
| openai → openai | growth_rate_mib_per_min | 0.37583951307297037 | valid | — |
| openai → openai | load_s | 300 | valid | — |
| openai → openai | shape | 0 | valid | — |
| openai → openai | idle_shape | -1 | valid | — |
| openai → openai | idle_static | 0 | valid | — |
| openai → openai | idle_growth_rate_mib_per_min | -0.8534572049737088 | valid | — |
| openai → openai | throughput_rps | 28888 req/s | concurrency 16; samples 173346 | 1 |
| openai → openai | throughput_rps | 31374 req/s | concurrency 64; samples 188254 | 5 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 10 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 50 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | 100 |
| openai → openai | throughput_rps | 32092 req/s | concurrency 128; samples 192583 | unbounded |
| openai-responses → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| bedrock → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| gemini → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| cohere → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → cohere | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → gemini | added_latency_p50_us | 162 | valid; gateway_c1_samples=32091; direct_c1_samples=257892 | — |
| anthropic → gemini | added_latency_p99_us | 187 | valid; gateway_c1_samples=32091; direct_c1_samples=257892 | — |
| anthropic → gemini | gateway_c1_p99_us | 216 | valid; gateway_c1_samples=32091 | — |
| anthropic → gemini | direct_c1_p99_us | 29 | valid; direct_c1_samples=257892 | — |
| anthropic → gemini | cpu_us_per_request | 163.12134905764356 | valid | — |
| anthropic → gemini | rps_per_cpu_second | 6130.405405405406 | valid | — |
| anthropic → gemini | cost_window_conc | 8 | valid | — |
| anthropic → gemini | cost_core_utilisation | 0.9835725913841037 | valid | — |
| anthropic → gemini | cost_window_ok | 145168 | valid | — |
| anthropic → gemini | cost_window_rps | 24191 | valid | — |
| anthropic → gemini | cost_threads | 16 | valid | — |
| anthropic → gemini | cost_nonvol_ctxt_per_request | 0 | valid | — |
| anthropic → gemini | cost_majflt | 0 | valid | — |
| anthropic → gemini | added_ttft_p50_us | untestable | untestable | — |
| anthropic → gemini | added_ttft_p99_us | untestable | untestable | — |
| anthropic → gemini | added_gap_p50_us | untestable | untestable | — |
| anthropic → gemini | added_gap_p99_us | untestable | untestable | — |
| anthropic → gemini | streams_sustained_fps | untestable | untestable | — |
| anthropic → gemini | streams_sustained | untestable | untestable | — |
| anthropic → gemini | idle_rss_mib | 34.71875 | valid | — |
| anthropic → gemini | steady_state_rss_mib | 67.51953125 | valid | — |
| anthropic → gemini | recovered_rss_mib | 52.1015625 | valid | — |
| anthropic → gemini | peak_rss_mib | 69.29296875 | valid | — |
| anthropic → gemini | peak_rss_hwm_mib | 69.109375 | valid | — |
| anthropic → gemini | time_to_plateau_s | not_measured | not_measured | — |
| anthropic → gemini | growth_rate_mib_per_min | 0.6152929252623993 | valid | — |
| anthropic → gemini | load_s | 300 | valid | — |
| anthropic → gemini | shape | 0 | valid | — |
| anthropic → gemini | idle_shape | -1 | valid | — |
| anthropic → gemini | idle_static | 0 | valid | — |
| anthropic → gemini | idle_growth_rate_mib_per_min | -0.7567316178676103 | valid | — |
| anthropic → gemini | throughput_rps | 26504 req/s | concurrency 16; samples 159044 | 1 |
| anthropic → gemini | throughput_rps | 28918 req/s | concurrency 64; samples 173514 | 5 |
| anthropic → gemini | throughput_rps | 29709 req/s | concurrency 128; samples 178283 | 10 |
| anthropic → gemini | throughput_rps | 29709 req/s | concurrency 128; samples 178283 | 50 |
| anthropic → gemini | throughput_rps | 29709 req/s | concurrency 128; samples 178283 | 100 |
| anthropic → gemini | throughput_rps | 29709 req/s | concurrency 128; samples 178283 | unbounded |
| bedrock → gemini | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| gemini → gemini | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai → gemini | added_latency_p50_us | 163 | valid; gateway_c1_samples=31889; direct_c1_samples=261875 | — |
| openai → gemini | added_latency_p99_us | 191 | valid; gateway_c1_samples=31889; direct_c1_samples=261875 | — |
| openai → gemini | gateway_c1_p99_us | 219 | valid; gateway_c1_samples=31889 | — |
| openai → gemini | direct_c1_p99_us | 28 | valid; direct_c1_samples=261875 | — |
| openai → gemini | cpu_us_per_request | 159.98538822669894 | valid | — |
| openai → gemini | rps_per_cpu_second | 6250.570824524313 | valid | — |
| openai → gemini | cost_window_conc | 8 | valid | — |
| openai → gemini | cost_core_utilisation | 0.98230796385704 | valid | — |
| openai → gemini | cost_window_ok | 147826 | valid | — |
| openai → gemini | cost_window_rps | 24634 | valid | — |
| openai → gemini | cost_threads | 16 | valid | — |
| openai → gemini | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → gemini | cost_majflt | 0 | valid | — |
| openai → gemini | added_ttft_p50_us | untestable | untestable | — |
| openai → gemini | added_ttft_p99_us | untestable | untestable | — |
| openai → gemini | added_gap_p50_us | untestable | untestable | — |
| openai → gemini | added_gap_p99_us | untestable | untestable | — |
| openai → gemini | streams_sustained_fps | untestable | untestable | — |
| openai → gemini | streams_sustained | untestable | untestable | — |
| openai → gemini | idle_rss_mib | 34.7265625 | valid | — |
| openai → gemini | steady_state_rss_mib | 67.9921875 | valid | — |
| openai → gemini | recovered_rss_mib | 53.26953125 | valid | — |
| openai → gemini | peak_rss_mib | 69.94921875 | valid | — |
| openai → gemini | peak_rss_hwm_mib | 69.70703125 | valid | — |
| openai → gemini | time_to_plateau_s | not_measured | not_measured | — |
| openai → gemini | growth_rate_mib_per_min | 2.0595721727593674 | valid | — |
| openai → gemini | load_s | 300 | valid | — |
| openai → gemini | shape | 1 | valid | — |
| openai → gemini | idle_shape | not_measured | not_measured | — |
| openai → gemini | idle_static | 1 | valid | — |
| openai → gemini | idle_growth_rate_mib_per_min | -0.4251936813206195 | valid | — |
| openai → gemini | throughput_rps | 24347 req/s | concurrency 8; samples 146100 | 1 |
| openai → gemini | throughput_rps | 29193 req/s | concurrency 64; samples 175189 | 5 |
| openai → gemini | throughput_rps | 29862 req/s | concurrency 128; samples 179198 | 10 |
| openai → gemini | throughput_rps | 29862 req/s | concurrency 128; samples 179198 | 50 |
| openai → gemini | throughput_rps | 29862 req/s | concurrency 128; samples 179198 | 100 |
| openai → gemini | throughput_rps | 29862 req/s | concurrency 128; samples 179198 | unbounded |
| openai-responses → gemini | added_latency_p50_us | 179 | valid; gateway_c1_samples=29315; direct_c1_samples=264153 | — |
| openai-responses → gemini | added_latency_p99_us | 209 | valid; gateway_c1_samples=29315; direct_c1_samples=264153 | — |
| openai-responses → gemini | gateway_c1_p99_us | 237 | valid; gateway_c1_samples=29315 | — |
| openai-responses → gemini | direct_c1_p99_us | 28 | valid; direct_c1_samples=264153 | — |
| openai-responses → gemini | cpu_us_per_request | 173.738380787614 | valid | — |
| openai-responses → gemini | rps_per_cpu_second | 5755.7805907173 | valid | — |
| openai-responses → gemini | cost_window_conc | 8 | valid | — |
| openai-responses → gemini | cost_core_utilisation | 0.9843873210247157 | valid | — |
| openai-responses → gemini | cost_window_ok | 136412 | valid | — |
| openai-responses → gemini | cost_window_rps | 22732 | valid | — |
| openai-responses → gemini | cost_threads | 16 | valid | — |
| openai-responses → gemini | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai-responses → gemini | cost_majflt | 0 | valid | — |
| openai-responses → gemini | added_ttft_p50_us | untestable | untestable | — |
| openai-responses → gemini | added_ttft_p99_us | untestable | untestable | — |
| openai-responses → gemini | added_gap_p50_us | untestable | untestable | — |
| openai-responses → gemini | added_gap_p99_us | untestable | untestable | — |
| openai-responses → gemini | streams_sustained_fps | untestable | untestable | — |
| openai-responses → gemini | streams_sustained | untestable | untestable | — |
| openai-responses → gemini | idle_rss_mib | 34.3359375 | valid | — |
| openai-responses → gemini | steady_state_rss_mib | 67.0234375 | valid | — |
| openai-responses → gemini | recovered_rss_mib | 52.6640625 | valid | — |
| openai-responses → gemini | peak_rss_mib | 68.56640625 | valid | — |
| openai-responses → gemini | peak_rss_hwm_mib | 68.2265625 | valid | — |
| openai-responses → gemini | time_to_plateau_s | not_measured | not_measured | — |
| openai-responses → gemini | growth_rate_mib_per_min | -0.08107899161353596 | valid | — |
| openai-responses → gemini | load_s | 300 | valid | — |
| openai-responses → gemini | shape | 0 | valid | — |
| openai-responses → gemini | idle_shape | -1 | valid | — |
| openai-responses → gemini | idle_static | 0 | valid | — |
| openai-responses → gemini | idle_growth_rate_mib_per_min | -0.8728974207168577 | valid | — |
| openai-responses → gemini | throughput_rps | 24971 req/s | concurrency 16; samples 149843 | 1 |
| openai-responses → gemini | throughput_rps | 27134 req/s | concurrency 64; samples 162810 | 5 |
| openai-responses → gemini | throughput_rps | 27902 req/s | concurrency 128; samples 167436 | 10 |
| openai-responses → gemini | throughput_rps | 27902 req/s | concurrency 128; samples 167436 | 50 |
| openai-responses → gemini | throughput_rps | 27902 req/s | concurrency 128; samples 167436 | 100 |
| openai-responses → gemini | throughput_rps | 27902 req/s | concurrency 128; samples 167436 | unbounded |
| cohere → gemini | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai-responses → bedrock | added_latency_p50_us | 408 | valid; gateway_c1_samples=13815; direct_c1_samples=263299 | — |
| openai-responses → bedrock | added_latency_p99_us | 458 | valid; gateway_c1_samples=13815; direct_c1_samples=263299 | — |
| openai-responses → bedrock | gateway_c1_p99_us | 486 | valid; gateway_c1_samples=13815 | — |
| openai-responses → bedrock | direct_c1_p99_us | 28 | valid; direct_c1_samples=263299 | — |
| openai-responses → bedrock | cpu_us_per_request | 417.1148158107018 | valid | — |
| openai-responses → bedrock | rps_per_cpu_second | 2397.4214343271556 | valid | — |
| openai-responses → bedrock | cost_window_conc | 8 | valid | — |
| openai-responses → bedrock | cost_core_utilisation | 0.5155080086482076 | valid | — |
| openai-responses → bedrock | cost_window_ok | 29752 | valid | — |
| openai-responses → bedrock | cost_window_rps | 4958 | valid | — |
| openai-responses → bedrock | cost_threads | 16 | valid | — |
| openai-responses → bedrock | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai-responses → bedrock | cost_majflt | 0 | valid | — |
| openai-responses → bedrock | added_ttft_p50_us | untestable | untestable | — |
| openai-responses → bedrock | added_ttft_p99_us | untestable | untestable | — |
| openai-responses → bedrock | added_gap_p50_us | untestable | untestable | — |
| openai-responses → bedrock | added_gap_p99_us | untestable | untestable | — |
| openai-responses → bedrock | streams_sustained_fps | untestable | untestable | — |
| openai-responses → bedrock | streams_sustained | untestable | untestable | — |
| openai-responses → bedrock | idle_rss_mib | 34.140625 | valid | — |
| openai-responses → bedrock | steady_state_rss_mib | 63.89453125 | valid | — |
| openai-responses → bedrock | recovered_rss_mib | 57.609375 | valid | — |
| openai-responses → bedrock | peak_rss_mib | 66.1484375 | valid | — |
| openai-responses → bedrock | peak_rss_hwm_mib | 65.8671875 | valid | — |
| openai-responses → bedrock | time_to_plateau_s | not_measured | not_measured | — |
| openai-responses → bedrock | growth_rate_mib_per_min | 2.1686498706373123 | valid | — |
| openai-responses → bedrock | load_s | 300 | valid | — |
| openai-responses → bedrock | shape | 1 | valid | — |
| openai-responses → bedrock | idle_shape | -1 | valid | — |
| openai-responses → bedrock | idle_static | 0 | valid | — |
| openai-responses → bedrock | idle_growth_rate_mib_per_min | -0.7747356627383634 | valid | — |
| openai-responses → bedrock | throughput_rps | 6806 req/s | concurrency 4; samples 40846 | 1 |
| openai-responses → bedrock | throughput_rps | 8726 req/s | concurrency 16; samples 52362 | 5 |
| openai-responses → bedrock | throughput_rps | 9937 req/s | concurrency 32; samples 59631 | 10 |
| openai-responses → bedrock | throughput_rps | 10099 req/s | concurrency 64; samples 60598 | 50 |
| openai-responses → bedrock | throughput_rps | 10099 req/s | concurrency 64; samples 60598 | 100 |
| openai-responses → bedrock | throughput_rps | 10099 req/s | concurrency 64; samples 60598 | unbounded |
| gemini → bedrock | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| bedrock → bedrock | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai → bedrock | added_latency_p50_us | 381 | valid; gateway_c1_samples=14735; direct_c1_samples=261517 | — |
| openai → bedrock | added_latency_p99_us | 427 | valid; gateway_c1_samples=14735; direct_c1_samples=261517 | — |
| openai → bedrock | gateway_c1_p99_us | 455 | valid; gateway_c1_samples=14735 | — |
| openai → bedrock | direct_c1_p99_us | 28 | valid; direct_c1_samples=261517 | — |
| openai → bedrock | cpu_us_per_request | 387.2113676731794 | valid | — |
| openai → bedrock | rps_per_cpu_second | 2582.5688073394494 | valid | — |
| openai → bedrock | cost_window_conc | 8 | valid | — |
| openai → bedrock | cost_core_utilisation | 0.5886832786469611 | valid | — |
| openai → bedrock | cost_window_ok | 36595 | valid | — |
| openai → bedrock | cost_window_rps | 6098 | valid | — |
| openai → bedrock | cost_threads | 16 | valid | — |
| openai → bedrock | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → bedrock | cost_majflt | 0 | valid | — |
| openai → bedrock | added_ttft_p50_us | untestable | untestable | — |
| openai → bedrock | added_ttft_p99_us | untestable | untestable | — |
| openai → bedrock | added_gap_p50_us | untestable | untestable | — |
| openai → bedrock | added_gap_p99_us | untestable | untestable | — |
| openai → bedrock | streams_sustained_fps | untestable | untestable | — |
| openai → bedrock | streams_sustained | untestable | untestable | — |
| openai → bedrock | idle_rss_mib | 34.74609375 | valid | — |
| openai → bedrock | steady_state_rss_mib | 63.51171875 | valid | — |
| openai → bedrock | recovered_rss_mib | 56.62109375 | valid | — |
| openai → bedrock | peak_rss_mib | 64.6875 | valid | — |
| openai → bedrock | peak_rss_hwm_mib | 64.42578125 | valid | — |
| openai → bedrock | time_to_plateau_s | not_measured | not_measured | — |
| openai → bedrock | growth_rate_mib_per_min | 0.34857051611065903 | valid | — |
| openai → bedrock | load_s | 300 | valid | — |
| openai → bedrock | shape | 0 | valid | — |
| openai → bedrock | idle_shape | not_measured | not_measured | — |
| openai → bedrock | idle_static | 1 | valid | — |
| openai → bedrock | idle_growth_rate_mib_per_min | -0.4981697903891496 | valid | — |
| openai → bedrock | throughput_rps | 4228 req/s | concurrency 2; samples 25376 | 1 |
| openai → bedrock | throughput_rps | 10244 req/s | concurrency 16; samples 61474 | 5 |
| openai → bedrock | throughput_rps | 10358 req/s | concurrency 32; samples 62160 | 10 |
| openai → bedrock | throughput_rps | 10529 req/s | concurrency 64; samples 63182 | 50 |
| openai → bedrock | throughput_rps | 10529 req/s | concurrency 64; samples 63182 | 100 |
| openai → bedrock | throughput_rps | 10529 req/s | concurrency 64; samples 63182 | unbounded |
| cohere → bedrock | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → bedrock | added_latency_p50_us | 400 | valid; gateway_c1_samples=14087; direct_c1_samples=261607 | — |
| anthropic → bedrock | added_latency_p99_us | 448 | valid; gateway_c1_samples=14087; direct_c1_samples=261607 | — |
| anthropic → bedrock | gateway_c1_p99_us | 476 | valid; gateway_c1_samples=14087 | — |
| anthropic → bedrock | direct_c1_p99_us | 28 | valid; direct_c1_samples=261607 | — |
| anthropic → bedrock | cpu_us_per_request | 389.410061071586 | valid | — |
| anthropic → bedrock | rps_per_cpu_second | 2567.9870654810024 | valid | — |
| anthropic → bedrock | cost_window_conc | 8 | valid | — |
| anthropic → bedrock | cost_core_utilisation | 0.5138925384814116 | valid | — |
| anthropic → bedrock | cost_window_ok | 31766 | valid | — |
| anthropic → bedrock | cost_window_rps | 5293 | valid | — |
| anthropic → bedrock | cost_threads | 16 | valid | — |
| anthropic → bedrock | cost_nonvol_ctxt_per_request | 0 | valid | — |
| anthropic → bedrock | cost_majflt | 0 | valid | — |
| anthropic → bedrock | added_ttft_p50_us | untestable | untestable | — |
| anthropic → bedrock | added_ttft_p99_us | untestable | untestable | — |
| anthropic → bedrock | added_gap_p50_us | untestable | untestable | — |
| anthropic → bedrock | added_gap_p99_us | untestable | untestable | — |
| anthropic → bedrock | streams_sustained_fps | untestable | untestable | — |
| anthropic → bedrock | streams_sustained | untestable | untestable | — |
| anthropic → bedrock | idle_rss_mib | 34.20703125 | valid | — |
| anthropic → bedrock | steady_state_rss_mib | 62.880859375 | valid | — |
| anthropic → bedrock | recovered_rss_mib | 56.078125 | valid | — |
| anthropic → bedrock | peak_rss_mib | 64.55078125 | valid | — |
| anthropic → bedrock | peak_rss_hwm_mib | 64.265625 | valid | — |
| anthropic → bedrock | time_to_plateau_s | not_measured | not_measured | — |
| anthropic → bedrock | growth_rate_mib_per_min | 1.520035788472686 | valid | — |
| anthropic → bedrock | load_s | 300 | valid | — |
| anthropic → bedrock | shape | 1 | valid | — |
| anthropic → bedrock | idle_shape | not_measured | not_measured | — |
| anthropic → bedrock | idle_static | 1 | valid | — |
| anthropic → bedrock | idle_growth_rate_mib_per_min | -0.40941641695154907 | valid | — |
| anthropic → bedrock | throughput_rps | 5623 req/s | concurrency 4; samples 33749 | 1 |
| anthropic → bedrock | throughput_rps | 10290 req/s | concurrency 32; samples 61750 | 5 |
| anthropic → bedrock | throughput_rps | 10340 req/s | concurrency 64; samples 62052 | 10 |
| anthropic → bedrock | throughput_rps | 10340 req/s | concurrency 64; samples 62052 | 50 |
| anthropic → bedrock | throughput_rps | 10340 req/s | concurrency 64; samples 62052 | 100 |
| anthropic → bedrock | throughput_rps | 10340 req/s | concurrency 64; samples 62052 | unbounded |
| gemini → openai-responses | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| cohere → openai-responses | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| bedrock → openai-responses | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| anthropic → openai-responses | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |
| openai-responses → openai-responses | added_latency_p50_us | 160 | valid; gateway_c1_samples=32575; direct_c1_samples=260507 | — |
| openai-responses → openai-responses | added_latency_p99_us | 188 | valid; gateway_c1_samples=32575; direct_c1_samples=260507 | — |
| openai-responses → openai-responses | gateway_c1_p99_us | 217 | valid; gateway_c1_samples=32575 | — |
| openai-responses → openai-responses | direct_c1_p99_us | 29 | valid; direct_c1_samples=260507 | — |
| openai-responses → openai-responses | cpu_us_per_request | 148.0885877329253 | valid | — |
| openai-responses → openai-responses | rps_per_cpu_second | 6752.714812862415 | valid | — |
| openai-responses → openai-responses | cost_window_conc | 8 | valid | — |
| openai-responses → openai-responses | cost_core_utilisation | 0.7879501484182734 | valid | — |
| openai-responses → openai-responses | cost_window_ok | 128099 | valid | — |
| openai-responses → openai-responses | cost_window_rps | 21347 | valid | — |
| openai-responses → openai-responses | cost_threads | 16 | valid | — |
| openai-responses → openai-responses | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai-responses → openai-responses | cost_majflt | 0 | valid | — |
| openai-responses → openai-responses | added_ttft_p50_us | untestable | untestable | — |
| openai-responses → openai-responses | added_ttft_p99_us | untestable | untestable | — |
| openai-responses → openai-responses | added_gap_p50_us | untestable | untestable | — |
| openai-responses → openai-responses | added_gap_p99_us | untestable | untestable | — |
| openai-responses → openai-responses | streams_sustained_fps | untestable | untestable | — |
| openai-responses → openai-responses | streams_sustained | untestable | untestable | — |
| openai-responses → openai-responses | idle_rss_mib | 34.328125 | valid | — |
| openai-responses → openai-responses | steady_state_rss_mib | 67.34765625 | valid | — |
| openai-responses → openai-responses | recovered_rss_mib | 51.8984375 | valid | — |
| openai-responses → openai-responses | peak_rss_mib | 69.07421875 | valid | — |
| openai-responses → openai-responses | peak_rss_hwm_mib | 68.703125 | valid | — |
| openai-responses → openai-responses | time_to_plateau_s | not_measured | not_measured | — |
| openai-responses → openai-responses | growth_rate_mib_per_min | -0.6509917696949409 | valid | — |
| openai-responses → openai-responses | load_s | 300 | valid | — |
| openai-responses → openai-responses | shape | 0 | valid | — |
| openai-responses → openai-responses | idle_shape | -1 | valid | — |
| openai-responses → openai-responses | idle_static | 0 | valid | — |
| openai-responses → openai-responses | idle_growth_rate_mib_per_min | -0.8039072804175704 | valid | — |
| openai-responses → openai-responses | throughput_rps | 25391 req/s | concurrency 8; samples 152368 | 1 |
| openai-responses → openai-responses | throughput_rps | 30114 req/s | concurrency 64; samples 180699 | 5 |
| openai-responses → openai-responses | throughput_rps | 31044 req/s | concurrency 128; samples 186270 | 10 |
| openai-responses → openai-responses | throughput_rps | 31044 req/s | concurrency 128; samples 186270 | 50 |
| openai-responses → openai-responses | throughput_rps | 31044 req/s | concurrency 128; samples 186270 | 100 |
| openai-responses → openai-responses | throughput_rps | 31044 req/s | concurrency 128; samples 186270 | unbounded |
| openai → openai-responses | — | — | not_declared: Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json | — |

## Probe evidence and instrument provenance

Historical capability notes may reference an earlier product version. They are not a new probe or a current support claim. Full RSS samples remain in the immutable raw JSON.

| Ingress → upstream | Cell state | Probe HTTP status | Recorded reason | Recorded detail | Evidence scope | Response excerpt | Tested build |
| --- | --- | --- | --- | --- | --- | --- | --- |
| openai → anthropic | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → anthropic | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → anthropic | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → anthropic | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → anthropic | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → anthropic | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → openai | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → openai | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → openai | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → openai | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → openai | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai → openai | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → cohere | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → gemini | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → gemini | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → gemini | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai → gemini | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → gemini | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → gemini | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → bedrock | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → bedrock | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → bedrock | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai → bedrock | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → bedrock | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → bedrock | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| gemini → openai-responses | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| cohere → openai-responses | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| bedrock → openai-responses | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| anthropic → openai-responses | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai-responses → openai-responses | served | — | — | — | Recorded run evidence | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |
| openai → openai-responses | not_declared | — | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 1](https://onthebench.ai/#note-1) | [recorded note 2](https://onthebench.ai/#note-2) | — | ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a |

**Harness commit**: `d240badd966933fa19eec0606b2d1f3945f18ad2`

**Mock SHA-256**: `99d01f709d8d9c1f49055aa265d5eb7475264d6b513911fe317aee74a71190f8`

**Load generator SHA-256**: `not recorded`

**somaxconn**: 65535

**tcp_max_syn_backlog**: 65535

## Stream sweep diagnostics

Per streaming-sweep concurrency, as recorded by the engine: lanes by final HTTP status, and lanes that received no response.

| Ingress → upstream | Concurrency | Lanes by HTTP status | Lanes with no response |
| --- | --- | --- | --- |
| openai → anthropic | 1 | {"200":1} | 0 |
| openai → anthropic | 2 | {"200":2} | 0 |
| openai → anthropic | 4 | {"200":4} | 0 |
| openai → anthropic | 8 | {"200":8} | 0 |
| openai → anthropic | 16 | {"200":16} | 0 |
| openai → anthropic | 32 | {"200":32} | 0 |
| openai → anthropic | 64 | {"200":64} | 0 |
| openai → anthropic | 128 | {"200":128} | 0 |
| openai → anthropic | 256 | {"200":256} | 0 |
| openai → anthropic | 512 | {"200":512} | 0 |
| openai → anthropic | 1024 | {"200":1024} | 0 |
| openai → anthropic | 768 | {"200":768} | 0 |
| openai → anthropic | 896 | {"200":896} | 0 |
| openai → anthropic | 832 | {"200":832} | 0 |
| openai → anthropic | 800 | {"200":800} | 0 |
| openai → anthropic | 784 | {"200":784} | 0 |
| openai → anthropic | 776 | {"200":776} | 0 |
| openai → anthropic | 772 | {"200":772} | 0 |
| openai → anthropic | 774 | {"200":774} | 0 |
| openai → anthropic | 775 | {"200":775} | 0 |
| openai → anthropic | 775 | {"200":775} | 0 |
| openai → anthropic | 775 | {"200":775} | 0 |
| anthropic → anthropic | 1 | {"200":1} | 0 |
| anthropic → anthropic | 2 | {"200":2} | 0 |
| anthropic → anthropic | 4 | {"200":4} | 0 |
| anthropic → anthropic | 8 | {"200":8} | 0 |
| anthropic → anthropic | 16 | {"200":16} | 0 |
| anthropic → anthropic | 32 | {"200":32} | 0 |
| anthropic → anthropic | 64 | {"200":64} | 0 |
| anthropic → anthropic | 128 | {"200":128} | 0 |
| anthropic → anthropic | 256 | {"200":256} | 0 |
| anthropic → anthropic | 512 | {"200":512} | 0 |
| anthropic → anthropic | 1024 | {"200":1024} | 0 |
| anthropic → anthropic | 768 | {"200":768} | 0 |
| anthropic → anthropic | 896 | {"200":896} | 0 |
| anthropic → anthropic | 832 | {"200":832} | 0 |
| anthropic → anthropic | 864 | {"200":864} | 0 |
| anthropic → anthropic | 848 | {"200":848} | 0 |
| anthropic → anthropic | 840 | {"200":840} | 0 |
| anthropic → anthropic | 844 | {"200":844} | 0 |
| anthropic → anthropic | 842 | {"200":842} | 0 |
| anthropic → anthropic | 843 | {"200":843} | 0 |
| anthropic → anthropic | 843 | {"200":843} | 0 |
| anthropic → anthropic | 843 | {"200":843} | 0 |
| anthropic → anthropic | 677 | {"200":677} | 0 |
| anthropic → anthropic | 677 | {"200":677} | 0 |
| anthropic → anthropic | 677 | {"200":677} | 0 |
| anthropic → openai | 1 | {"200":1} | 0 |
| anthropic → openai | 2 | {"200":2} | 0 |
| anthropic → openai | 4 | {"200":4} | 0 |
| anthropic → openai | 8 | {"200":8} | 0 |
| anthropic → openai | 16 | {"200":16} | 0 |
| anthropic → openai | 32 | {"200":32} | 0 |
| anthropic → openai | 64 | {"200":64} | 0 |
| anthropic → openai | 128 | {"200":128} | 0 |
| anthropic → openai | 256 | {"200":256} | 0 |
| anthropic → openai | 512 | {"200":512} | 0 |
| anthropic → openai | 1024 | {"200":1024} | 0 |
| anthropic → openai | 768 | {"200":768} | 0 |
| anthropic → openai | 896 | {"200":896} | 0 |
| anthropic → openai | 832 | {"200":832} | 0 |
| anthropic → openai | 800 | {"200":800} | 0 |
| anthropic → openai | 816 | {"200":816} | 0 |
| anthropic → openai | 808 | {"200":808} | 0 |
| anthropic → openai | 804 | {"200":804} | 0 |
| anthropic → openai | 806 | {"200":806} | 0 |
| anthropic → openai | 805 | {"200":805} | 0 |
| anthropic → openai | 804 | {"200":804} | 0 |
| anthropic → openai | 804 | {"200":804} | 0 |
| anthropic → openai | 658 | {"200":658} | 0 |
| anthropic → openai | 658 | {"200":658} | 0 |
| anthropic → openai | 658 | {"200":658} | 0 |
| openai → openai | 1 | {"200":1} | 0 |
| openai → openai | 2 | {"200":2} | 0 |
| openai → openai | 4 | {"200":4} | 0 |
| openai → openai | 8 | {"200":8} | 0 |
| openai → openai | 16 | {"200":16} | 0 |
| openai → openai | 32 | {"200":32} | 0 |
| openai → openai | 64 | {"200":64} | 0 |
| openai → openai | 128 | {"200":128} | 0 |
| openai → openai | 256 | {"200":256} | 0 |
| openai → openai | 512 | {"200":512} | 0 |
| openai → openai | 1024 | {"200":1024} | 0 |
| openai → openai | 768 | {"200":768} | 0 |
| openai → openai | 896 | {"200":896} | 0 |
| openai → openai | 832 | {"200":832} | 0 |
| openai → openai | 800 | {"200":800} | 0 |
| openai → openai | 784 | {"200":784} | 0 |
| openai → openai | 776 | {"200":776} | 0 |
| openai → openai | 780 | {"200":780} | 0 |
| openai → openai | 778 | {"200":778} | 0 |
| openai → openai | 779 | {"200":779} | 0 |
| openai → openai | 779 | {"200":779} | 0 |
| openai → openai | 779 | {"200":779} | 0 |
| openai → openai | 645 | {"200":645} | 0 |
| openai → openai | 645 | {"200":645} | 0 |
| openai → openai | 645 | {"200":645} | 0 |

## Recorded notes

Text recorded with the cells above. Each note is shown once; rows refer to it by number.

- Pinned AISIX v1.9.0 source 7e070f0bde3f653c3b197ed18c5b9205b866a4c5: README native adapter families, aisix-proxy/src/lib.rs routes and dispatch.rs::serves_natively_with, provider-vertex/bedrock bridge configuration. OpenAI, Responses and Anthropic ingress translate to Chat Completions, Anthropic, Gemini and Bedrock; only Responses ingress selects the Responses upstream. Separate provider keys use explicit apis:{} for Chat-only routing and the native Responses declaration default for the Responses alias. Thirteen exact pairs passed local recorder checks. Evidence: results/diagnostics/refresh-configuration/aisix-protocol-preflight.json
- Historical capability research, not a fresh probe of the tested build. Referenced versions may differ from gateway_build; do not infer current capability.

## Evidence for the current measurements

- [run-aisix-ef0cff28391e0017f17c4b7c01e0171c62359468b787de48402bc3408df76915](https://onthebench.ai/gateways/runs/run-aisix-ef0cff28391e0017f17c4b7c01e0171c62359468b787de48402bc3408df76915)  ·  [complete raw record including RSS series](https://onthebench.ai/api/v1/runs/run-aisix-ef0cff28391e0017f17c4b7c01e0171c62359468b787de48402bc3408df76915/raw.json)

## Configuration recorded by the run

```
# config.gen.yaml
resources_file: "/etc/aisix/resources.yaml"
proxy:
  addr: "0.0.0.0:3000"
admin:
  admin_keys:
    - "aisix-admin-dummy"

# resources.gen.yaml
_format_version: "1"
provider_keys:
  - display_name: mock-openai
    provider: openai
    adapter: openai
    apis: {}
    api_base: "http://127.0.0.1:8000/v1"
    api_key: "sk-mock"
  - display_name: mock-anthropic
    provider: anthropic
    adapter: anthropic
    api_base: "http://127.0.0.1:8000"
    api_key: "sk-mock"
  - display_name: mock-vertex
    provider: google
    adapter: vertex
    api_base: "http://127.0.0.1:8000"
    api_key: '{"access_token":"dummy","project":"mock","region":"us-central1"}'
  - display_name: mock-bedrock
    provider: bedrock
    adapter: bedrock
    api_base: "http://127.0.0.1:8000"
    api_key: '{"access_key_id":"AKIAMOCKACCESSKEY","secret_access_key":"mock-secret","region":"us-east-1"}'
  - display_name: mock-responses
    provider: openai
    adapter: openai
    api_base: "http://127.0.0.1:8000/v1"
    api_key: "sk-mock"
models:
  - display_name: gpt-4o-mini
    provider: openai
    model_name: gpt-4o-mini
    provider_key: mock-openai
  - display_name: claude-3-5-sonnet-20241022
    provider: anthropic
    model_name: claude-3-5-sonnet-20241022
    provider_key: mock-anthropic
  - display_name: gemini-1.5-pro
    provider: google
    model_name: gemini-1.5-pro
    provider_key: mock-vertex
  - display_name: amazon.nova-micro-v1:0
    provider: bedrock
    model_name: amazon.nova-micro-v1:0
    provider_key: mock-bedrock
  - display_name: mock-responses
    provider: openai
    model_name: gpt-4o-mini
    provider_key: mock-responses
api_keys:
  - display_name: bench
    key_env: BENCH_AISIX_KEY
    allowed_models: ["*"]

```

## Compare a shared protocol cell

- [AISIX (api7) vs agentgateway](https://onthebench.ai/gateways/compare/aisix-vs-agentgateway)
- [AISIX (api7) vs APISIX](https://onthebench.ai/gateways/compare/aisix-vs-apisix)
- [AISIX (api7) vs Bifrost](https://onthebench.ai/gateways/compare/aisix-vs-bifrost)
- [AISIX (api7) vs Busbar](https://onthebench.ai/gateways/compare/aisix-vs-busbar)
- [AISIX (api7) vs GoModel](https://onthebench.ai/gateways/compare/aisix-vs-gomodel)
- [AISIX (api7) vs Helicone](https://onthebench.ai/gateways/compare/aisix-vs-helicone)
- [AISIX (api7) vs Higress](https://onthebench.ai/gateways/compare/aisix-vs-higress)
- [AISIX (api7) vs Kong](https://onthebench.ai/gateways/compare/aisix-vs-kong)
- [AISIX (api7) vs LiteLLM · Python](https://onthebench.ai/gateways/compare/aisix-vs-litellm-python)
- [AISIX (api7) vs LiteLLM · Rust](https://onthebench.ai/gateways/compare/aisix-vs-litellm-rust)
- [AISIX (api7) vs One-API](https://onthebench.ai/gateways/compare/aisix-vs-one-api)
- [AISIX (api7) vs Plano](https://onthebench.ai/gateways/compare/aisix-vs-plano)
- [AISIX (api7) vs Portkey](https://onthebench.ai/gateways/compare/aisix-vs-portkey)
- [AISIX (api7) vs TensorZero](https://onthebench.ai/gateways/compare/aisix-vs-tensorzero)

## Historical measurements

- [2026-11-10T17:39:23Z — ghcr.io/api7/aisix:1.9.0@sha256:dbe230bdaa4f353f7bf99d5437ab11f8b67435afabb217cf371c441ede43270a](https://onthebench.ai/gateways/runs/run-aisix-ef0cff28391e0017f17c4b7c01e0171c62359468b787de48402bc3408df76915)
- [2026-08-21T16:25:52Z — target/release/aisix](https://onthebench.ai/gateways/runs/run-aisix-2d9a80d1c05e9bae6d62c5eb1b7250ad64e98d8c56a7ab373cae7c512d249bda)
