# Helicone 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: target/release/ai-gateway. Measured: 2026-11-10T17:39:37Z.

[JSON](https://onthebench.ai/api/v1/gateways/helicone)  ·  [Markdown](https://onthebench.ai/gateways/helicone.md)  ·  [CSV](https://onthebench.ai/gateways/data.csv?gateway=helicone&compact=1)  ·  [Project source](https://github.com/Helicone/ai-gateway)  ·  [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) |
| --- | --- | --- | --- | --- |
| bedrock → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| gemini → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| cohere → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → cohere | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → openai | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → openai | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai → openai | added_latency_p50_us | 256 | valid; gateway_c1_samples=21418; direct_c1_samples=260967 | — |
| openai → openai | added_latency_p99_us | 274 | valid; gateway_c1_samples=21418; direct_c1_samples=260967 | — |
| openai → openai | gateway_c1_p99_us | 302 | valid; gateway_c1_samples=21418 | — |
| openai → openai | direct_c1_p99_us | 28 | valid; direct_c1_samples=260967 | — |
| openai → openai | cpu_us_per_request | 259.381777553837 | valid | — |
| openai → openai | rps_per_cpu_second | 3855.320945945946 | valid | — |
| openai → openai | cost_window_conc | 8 | valid | — |
| openai → openai | cost_core_utilisation | 0.9840922276022175 | valid | — |
| openai → openai | cost_window_ok | 91294 | valid | — |
| openai → openai | cost_window_rps | 15213 | valid | — |
| openai → openai | cost_threads | 24 | valid | — |
| openai → openai | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → openai | cost_majflt | 0 | valid | — |
| openai → openai | added_ttft_p50_us | 399 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_ttft_p99_us | 493 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → openai | added_gap_p50_us | 0 | 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 | 21883.075061106818 | valid | — |
| openai → openai | streams_sustained | 463 | valid | — |
| openai → openai | idle_rss_mib | 42.97265625 | valid | — |
| openai → openai | steady_state_rss_mib | 56.1796875 | valid | — |
| openai → openai | recovered_rss_mib | 53.69140625 | valid | — |
| openai → openai | peak_rss_mib | 56.94921875 | valid | — |
| openai → openai | peak_rss_hwm_mib | 56.609375 | valid | — |
| openai → openai | time_to_plateau_s | 126 | valid | — |
| openai → openai | growth_rate_mib_per_min | 0.4999152222931118 | valid | — |
| openai → openai | load_s | 300 | valid | — |
| openai → openai | shape | not_measured | not_measured | — |
| openai → openai | idle_shape | not_measured | not_measured | — |
| openai → openai | idle_static | 1 | valid | — |
| openai → openai | idle_growth_rate_mib_per_min | 0.12448848274528536 | valid | — |
| openai → openai | throughput_rps | 15405 req/s | concurrency 8; samples 92444 | 1 |
| openai → openai | throughput_rps | 16288 req/s | concurrency 32; samples 97741 | 5 |
| openai → openai | throughput_rps | 16376 req/s | concurrency 64; samples 98266 | 10 |
| openai → openai | throughput_rps | 16376 req/s | concurrency 64; samples 98266 | 50 |
| openai → openai | throughput_rps | 16376 req/s | concurrency 64; samples 98266 | 100 |
| openai → openai | throughput_rps | 16376 req/s | concurrency 64; samples 98266 | unbounded |
| cohere → openai | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| bedrock → openai | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| gemini → openai | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → anthropic | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| cohere → anthropic | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai → anthropic | added_latency_p50_us | 259 | valid; gateway_c1_samples=21200; direct_c1_samples=264063 | — |
| openai → anthropic | added_latency_p99_us | 278 | valid; gateway_c1_samples=21200; direct_c1_samples=264063 | — |
| openai → anthropic | gateway_c1_p99_us | 305 | valid; gateway_c1_samples=21200 | — |
| openai → anthropic | direct_c1_p99_us | 27 | valid; direct_c1_samples=264063 | — |
| openai → anthropic | cpu_us_per_request | 260.87723553928595 | valid | — |
| openai → anthropic | rps_per_cpu_second | 3833.2206255283177 | valid | — |
| openai → anthropic | cost_window_conc | 8 | valid | — |
| openai → anthropic | cost_core_utilisation | 0.9833009260344443 | valid | — |
| openai → anthropic | cost_window_ok | 90694 | valid | — |
| openai → anthropic | cost_window_rps | 15114 | valid | — |
| openai → anthropic | cost_threads | 26 | valid | — |
| openai → anthropic | cost_nonvol_ctxt_per_request | 0 | valid | — |
| openai → anthropic | cost_majflt | 0 | valid | — |
| openai → anthropic | added_ttft_p50_us | 400 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → anthropic | added_ttft_p99_us | 515 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| openai → anthropic | added_gap_p50_us | 15 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → anthropic | added_gap_p99_us | 20 | valid; gap_gw_samples=63; gap_direct_samples=63 | — |
| openai → anthropic | streams_sustained_fps | 22502.421498049265 | valid | — |
| openai → anthropic | streams_sustained | 477 | valid | — |
| openai → anthropic | idle_rss_mib | 42.8515625 | valid | — |
| openai → anthropic | steady_state_rss_mib | 56.48046875 | valid | — |
| openai → anthropic | recovered_rss_mib | 53.44140625 | valid | — |
| openai → anthropic | peak_rss_mib | 57.0390625 | valid | — |
| openai → anthropic | peak_rss_hwm_mib | 56.78515625 | valid | — |
| openai → anthropic | time_to_plateau_s | 168 | valid | — |
| openai → anthropic | growth_rate_mib_per_min | 0.10989598445016407 | valid | — |
| openai → anthropic | load_s | 300 | valid | — |
| openai → anthropic | shape | not_measured | not_measured | — |
| openai → anthropic | idle_shape | not_measured | not_measured | — |
| openai → anthropic | idle_static | 1 | valid | — |
| openai → anthropic | idle_growth_rate_mib_per_min | 0.1243941906234104 | valid | — |
| openai → anthropic | throughput_rps | 15212 req/s | concurrency 8; samples 91291 | 1 |
| openai → anthropic | throughput_rps | 16081 req/s | concurrency 32; samples 96498 | 5 |
| openai → anthropic | throughput_rps | 16164 req/s | concurrency 64; samples 97003 | 10 |
| openai → anthropic | throughput_rps | 16164 req/s | concurrency 64; samples 97003 | 50 |
| openai → anthropic | throughput_rps | 16164 req/s | concurrency 64; samples 97003 | 100 |
| openai → anthropic | throughput_rps | 16164 req/s | concurrency 64; samples 97003 | unbounded |
| bedrock → anthropic | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → anthropic | added_latency_p50_us | 175 | valid; gateway_c1_samples=30620; direct_c1_samples=260783 | — |
| anthropic → anthropic | added_latency_p99_us | 192 | valid; gateway_c1_samples=30620; direct_c1_samples=260783 | — |
| anthropic → anthropic | gateway_c1_p99_us | 220 | valid; gateway_c1_samples=30620 | — |
| anthropic → anthropic | direct_c1_p99_us | 28 | valid; direct_c1_samples=260783 | — |
| anthropic → anthropic | cpu_us_per_request | 166.79831861652607 | valid | — |
| anthropic → anthropic | rps_per_cpu_second | 5995.264270613108 | valid | — |
| anthropic → anthropic | cost_window_conc | 8 | valid | — |
| anthropic → anthropic | cost_core_utilisation | 0.9827635256920014 | valid | — |
| anthropic → anthropic | cost_window_ok | 141788 | valid | — |
| anthropic → anthropic | cost_window_rps | 23628 | valid | — |
| anthropic → anthropic | cost_threads | 5 | valid | — |
| anthropic → anthropic | cost_nonvol_ctxt_per_request | 0 | valid | — |
| anthropic → anthropic | cost_majflt | 0 | valid | — |
| anthropic → anthropic | added_ttft_p50_us | 270 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → anthropic | added_ttft_p99_us | 404 | valid; ttft_gw_samples=100; ttft_direct_samples=100 | — |
| anthropic → anthropic | added_gap_p50_us | below_resolution | below_resolution; 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 | 51526.561485962746 | valid | — |
| anthropic → anthropic | streams_sustained | 1142 | valid | — |
| anthropic → anthropic | idle_rss_mib | 42.9453125 | valid | — |
| anthropic → anthropic | steady_state_rss_mib | 50.80078125 | valid | — |
| anthropic → anthropic | recovered_rss_mib | 50.78515625 | valid | — |
| anthropic → anthropic | peak_rss_mib | 51.328125 | valid | — |
| anthropic → anthropic | peak_rss_hwm_mib | 51.1328125 | valid | — |
| anthropic → anthropic | time_to_plateau_s | not_measured | not_measured | — |
| anthropic → anthropic | growth_rate_mib_per_min | -0.21467345625353568 | 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 | 0.12427634769142537 | valid | — |
| anthropic → anthropic | throughput_rps | 24452 req/s | concurrency 16; samples 146729 | 1 |
| anthropic → anthropic | throughput_rps | 25909 req/s | concurrency 64; samples 155481 | 5 |
| anthropic → anthropic | throughput_rps | 26217 req/s | concurrency 128; samples 157328 | 10 |
| anthropic → anthropic | throughput_rps | 26320 req/s | concurrency 256; samples 157941 | 50 |
| anthropic → anthropic | throughput_rps | 26320 req/s | concurrency 256; samples 157941 | 100 |
| anthropic → anthropic | throughput_rps | 26320 req/s | concurrency 256; samples 157941 | unbounded |
| gemini → anthropic | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai → bedrock | added_latency_p50_us | 293 | valid; gateway_c1_samples=18960; direct_c1_samples=263027 | — |
| openai → bedrock | added_latency_p99_us | 311 | valid; gateway_c1_samples=18960; direct_c1_samples=263027 | — |
| openai → bedrock | gateway_c1_p99_us | 338 | valid; gateway_c1_samples=18960 | — |
| openai → bedrock | direct_c1_p99_us | 27 | valid; direct_c1_samples=263027 | — |
| openai → bedrock | cpu_us_per_request | 295.1100305157913 | valid | — |
| openai → bedrock | rps_per_cpu_second | 3388.5666246321985 | valid | — |
| openai → bedrock | cost_window_conc | 8 | valid | — |
| openai → bedrock | cost_core_utilisation | 0.9886809407528605 | valid | — |
| openai → bedrock | cost_window_ok | 80614 | valid | — |
| openai → bedrock | cost_window_rps | 13433 | valid | — |
| openai → bedrock | cost_threads | 19 | 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 | 42.97265625 | valid | — |
| openai → bedrock | steady_state_rss_mib | 56.4921875 | valid | — |
| openai → bedrock | recovered_rss_mib | 53.671875 | valid | — |
| openai → bedrock | peak_rss_mib | 56.91015625 | valid | — |
| openai → bedrock | peak_rss_hwm_mib | 56.7265625 | valid | — |
| openai → bedrock | time_to_plateau_s | 138 | valid | — |
| openai → bedrock | growth_rate_mib_per_min | 0.041694531242001856 | valid | — |
| openai → bedrock | load_s | 300 | valid | — |
| openai → bedrock | shape | not_measured | not_measured | — |
| openai → bedrock | idle_shape | not_measured | not_measured | — |
| openai → bedrock | idle_static | 1 | valid | — |
| openai → bedrock | idle_growth_rate_mib_per_min | 0.12441485584633757 | valid | — |
| openai → bedrock | throughput_rps | 13411 req/s | concurrency 8; samples 80477 | 1 |
| openai → bedrock | throughput_rps | 14128 req/s | concurrency 32; samples 84779 | 5 |
| openai → bedrock | throughput_rps | 14164 req/s | concurrency 64; samples 84985 | 10 |
| openai → bedrock | throughput_rps | 14164 req/s | concurrency 64; samples 84985 | 50 |
| openai → bedrock | throughput_rps | 14164 req/s | concurrency 64; samples 84985 | 100 |
| openai → bedrock | throughput_rps | 14164 req/s | concurrency 64; samples 84985 | unbounded |
| gemini → bedrock | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → bedrock | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| cohere → bedrock | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → bedrock | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| bedrock → bedrock | added_latency_p50_us | 203 | valid; gateway_c1_samples=26758; direct_c1_samples=262304 | — |
| bedrock → bedrock | added_latency_p99_us | 220 | valid; gateway_c1_samples=26758; direct_c1_samples=262304 | — |
| bedrock → bedrock | gateway_c1_p99_us | 247 | valid; gateway_c1_samples=26758 | — |
| bedrock → bedrock | direct_c1_p99_us | 27 | valid; direct_c1_samples=262304 | — |
| bedrock → bedrock | cpu_us_per_request | 191.48936170212767 | valid | — |
| bedrock → bedrock | rps_per_cpu_second | 5222.222222222222 | valid | — |
| bedrock → bedrock | cost_window_conc | 8 | valid | — |
| bedrock → bedrock | cost_core_utilisation | 0.9872755366427802 | valid | — |
| bedrock → bedrock | cost_window_ok | 124080 | valid | — |
| bedrock → bedrock | cost_window_rps | 20677 | valid | — |
| bedrock → bedrock | cost_threads | 5 | valid | — |
| bedrock → bedrock | cost_nonvol_ctxt_per_request | 0.000008059316569954868 | valid | — |
| bedrock → bedrock | cost_majflt | 0 | valid | — |
| bedrock → bedrock | added_ttft_p50_us | untestable | untestable | — |
| bedrock → bedrock | added_ttft_p99_us | untestable | untestable | — |
| bedrock → bedrock | added_gap_p50_us | untestable | untestable | — |
| bedrock → bedrock | added_gap_p99_us | untestable | untestable | — |
| bedrock → bedrock | streams_sustained_fps | untestable | untestable | — |
| bedrock → bedrock | streams_sustained | untestable | untestable | — |
| bedrock → bedrock | idle_rss_mib | 42.9609375 | valid | — |
| bedrock → bedrock | steady_state_rss_mib | 50.83203125 | valid | — |
| bedrock → bedrock | recovered_rss_mib | 51.14453125 | valid | — |
| bedrock → bedrock | peak_rss_mib | 51.4375 | valid | — |
| bedrock → bedrock | peak_rss_hwm_mib | 51.44140625 | valid | — |
| bedrock → bedrock | time_to_plateau_s | not_measured | not_measured | — |
| bedrock → bedrock | growth_rate_mib_per_min | -0.0079210740835309 | valid | — |
| bedrock → bedrock | load_s | 300 | valid | — |
| bedrock → bedrock | shape | 0 | valid | — |
| bedrock → bedrock | idle_shape | not_measured | not_measured | — |
| bedrock → bedrock | idle_static | 1 | valid | — |
| bedrock → bedrock | idle_growth_rate_mib_per_min | 0.1244287498577021 | valid | — |
| bedrock → bedrock | throughput_rps | 20749 req/s | concurrency 8; samples 124513 | 1 |
| bedrock → bedrock | throughput_rps | 22637 req/s | concurrency 64; samples 135825 | 5 |
| bedrock → bedrock | throughput_rps | 22807 req/s | concurrency 128; samples 136867 | 10 |
| bedrock → bedrock | throughput_rps | 22894 req/s | concurrency 256; samples 137389 | 50 |
| bedrock → bedrock | throughput_rps | 22894 req/s | concurrency 256; samples 137389 | 100 |
| bedrock → bedrock | throughput_rps | 22894 req/s | concurrency 256; samples 137389 | unbounded |
| openai → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| gemini → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| cohere → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| bedrock → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → gemini | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| cohere → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| bedrock → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| anthropic → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| openai-responses → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-preflight.json | — |
| gemini → openai-responses | — | — | not_declared: Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-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 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| anthropic → 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) | — | target/release/ai-gateway |
| openai-responses → 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) | — | target/release/ai-gateway |
| openai → openai | served | — | — | — | Recorded run evidence | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai-responses → 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai → anthropic | served | — | — | — | Recorded run evidence | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| anthropic → anthropic | served | — | — | — | Recorded run evidence | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai → bedrock | served | — | — | — | Recorded run evidence | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai-responses → 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| anthropic → 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) | — | target/release/ai-gateway |
| bedrock → bedrock | served | — | — | — | Recorded run evidence | — | target/release/ai-gateway |
| openai → 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) | — | target/release/ai-gateway |
| anthropic → 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai-responses → 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |
| openai-responses → 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) | — | target/release/ai-gateway |
| 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) | — | target/release/ai-gateway |

**Harness commit**: `d240badd966933fa19eec0606b2d1f3945f18ad2`

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

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

**somaxconn**: 65535

**tcp_max_syn_backlog**: 65535

**gateway_provenance**: **binary_name**: ai-gateway

**binary_sha256**: `f401a719a6341a95be9d1f0b7a63ce385241583ee632f5724df3bf54b04a22f1`

**cargo_lock_sha256**: `d70600602afa1fa8a054251da84622cd764c920e110bf0e3266f0a562473de88`

**compiler**: rustc 1.99.1 (58cf8f9c3 2026-10-13) binary: rustc commit-hash: 58cf8f9c35b355199727627d07ae34a4e2846d67 commit-date: 2026-10-13 host: aarch64-unknown-linux-gnu release: 1.99.1 LLVM version: 23.1.1

**source_commit**: `081f310be779f27122e6038e81e2359c44db4ecb`

**source_dirty**: false

## 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 → 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":460,"404":52} | 0 |
| openai → openai | 384 | {"200":384} | 0 |
| openai → openai | 448 | {"200":448} | 0 |
| openai → openai | 480 | {"200":391,"404":89} | 0 |
| openai → openai | 464 | {"200":463,"404":1} | 0 |
| openai → openai | 456 | {"200":456} | 0 |
| openai → openai | 460 | {"200":460} | 0 |
| openai → openai | 462 | {"200":462} | 0 |
| openai → openai | 463 | {"200":463} | 0 |
| openai → openai | 463 | {"200":463} | 0 |
| openai → openai | 463 | {"200":463} | 0 |
| 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":511,"404":1} | 0 |
| openai → anthropic | 384 | {"200":384} | 0 |
| openai → anthropic | 448 | {"200":448} | 0 |
| openai → anthropic | 480 | {"200":463,"404":17} | 0 |
| openai → anthropic | 464 | {"200":464} | 0 |
| openai → anthropic | 472 | {"200":472} | 0 |
| openai → anthropic | 476 | {"200":476} | 0 |
| openai → anthropic | 478 | {"200":378,"404":100} | 0 |
| openai → anthropic | 477 | {"200":477} | 0 |
| openai → anthropic | 477 | {"200":477} | 0 |
| openai → anthropic | 477 | {"200":477} | 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 | 2048 | {"200":2048} | 0 |
| anthropic → anthropic | 1536 | {"200":1536} | 0 |
| anthropic → anthropic | 1280 | {"200":1280} | 0 |
| anthropic → anthropic | 1152 | {"200":1152} | 0 |
| anthropic → anthropic | 1088 | {"200":1088} | 0 |
| anthropic → anthropic | 1120 | {"200":1120} | 0 |
| anthropic → anthropic | 1136 | {"200":1136} | 0 |
| anthropic → anthropic | 1144 | {"200":1144} | 0 |
| anthropic → anthropic | 1140 | {"200":1140} | 0 |
| anthropic → anthropic | 1142 | {"200":1142} | 0 |
| anthropic → anthropic | 1143 | {"200":1143} | 0 |
| anthropic → anthropic | 1142 | {"200":1142} | 0 |
| anthropic → anthropic | 1142 | {"200":1142} | 0 |

## Recorded notes

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

- Pinned v0.2.0-beta.30 source 081f310be779f27122e6038e81e2359c44db4ecb: ai-gateway/src/endpoints defines OpenAI Chat, Anthropic Messages and Bedrock Converse. Google uses v1beta/openai/chat/completions, not native Gemini. router/router_details.rs exposes OpenAI router ingress plus provider-prefixed direct proxy paths. No Responses, native Gemini or Cohere interface is registered; direct proxy does not imply cross-protocol conversion. Five supported exact-wire pairs pass the 36-cell correctness check. Evidence: results/diagnostics/refresh-configuration/helicone-final-config-correctness-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-helicone-4a55ff2dcc509c5e03f85bd92e36b949ac410653128ccc77d60019bb3aaf2558](https://onthebench.ai/gateways/runs/run-helicone-4a55ff2dcc509c5e03f85bd92e36b949ac410653128ccc77d60019bb3aaf2558)  ·  [complete raw record including RSS series](https://onthebench.ai/api/v1/runs/run-helicone-4a55ff2dcc509c5e03f85bd92e36b949ac410653128ccc77d60019bb3aaf2558/raw.json)

## Configuration recorded by the run

```
server:
  port: 8787
providers:
  openai:
    base-url: "http://127.0.0.1:8000/"
  anthropic:
    base-url: "http://127.0.0.1:8000/"
  bedrock:
    base-url: "http://127.0.0.1:8000/"
routers:
  openai:
    load-balance:
      chat:
        strategy: latency
        providers:
          - openai
  anthropic:
    load-balance:
      chat:
        strategy: latency
        providers:
          - anthropic
  bedrock:
    load-balance:
      chat:
        strategy: latency
        providers:
          - bedrock

```

## Compare a shared protocol cell

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

## Historical measurements

- [2026-11-10T17:39:37Z — target/release/ai-gateway](https://onthebench.ai/gateways/runs/run-helicone-4a55ff2dcc509c5e03f85bd92e36b949ac410653128ccc77d60019bb3aaf2558)
- [2026-08-03T21:30:02Z — target/release/ai-gateway](https://onthebench.ai/gateways/runs/run-helicone-6ad9b8767fe745f1cf0967cbcfa2719c632a742aa5dc214104cc0affacb51548)
