Portkey 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.
Node · recorded build: portkeyai/gateway:1.15.2. Measured: .
JSON · Markdown · CSV · Project source · Interactive dashboard
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 → OpenAI | — | — | Not declared: recorded note 1 | — |
| Gemini → OpenAI | — | — | Not declared: recorded note 1 | — |
| OpenAI → OpenAI | Added latency p50 (µs) | 1,196 | Valid · n=4,304 gateway / 246,357 direct | — |
| OpenAI → OpenAI | Added latency p99 (µs) | 3,902 | Valid · n=4,304 gateway / 246,357 direct | — |
| OpenAI → OpenAI | Gateway p99 at concurrency 1 (µs) | 3,938 | Valid · n=4,304 | — |
| OpenAI → OpenAI | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=246,357 | — |
| OpenAI → OpenAI | CPU per request (µs) | 1,528.4 | Valid | — |
| OpenAI → OpenAI | Requests per CPU-second | 654.3 | Valid | — |
| OpenAI → OpenAI | Cost window concurrency | 8 | Valid | — |
| OpenAI → OpenAI | Cost window core utilisation | 0.295 | Valid | — |
| OpenAI → OpenAI | Cost window successful requests | 4,652 | Valid | — |
| OpenAI → OpenAI | Cost window throughput (req/s) | 775 | Valid | — |
| OpenAI → OpenAI | Cost window threads | 22 | Valid | — |
| OpenAI → OpenAI | Non-voluntary context switches per request | 0.007 | Valid | — |
| OpenAI → OpenAI | Cost window major page faults | 0 | Valid | — |
| OpenAI → OpenAI | Added time to first event p50 (µs) | 28,114 | Valid · n=100 gateway / 100 direct | — |
| OpenAI → OpenAI | Added time to first event p99 (µs) | 29,752 | Valid · n=100 gateway / 100 direct | — |
| OpenAI → OpenAI | Added inter-frame gap p50 (µs) | 0 | Valid · n=63 gateway / 63 direct | — |
| OpenAI → OpenAI | Added inter-frame gap p99 (µs) | 440 | Valid · n=63 gateway / 63 direct | — |
| OpenAI → OpenAI | Streams sustained (frames/s) | 4,996.2 | Valid | — |
| OpenAI → OpenAI | Streams sustained | 1,519 | Valid | — |
| OpenAI → OpenAI | Idle RSS (MiB) | 125.1 | Valid | — |
| OpenAI → OpenAI | Steady-state RSS (MiB) | 245.2 | Valid | — |
| OpenAI → OpenAI | Recovered RSS (MiB) | 142.1 | Valid | — |
| OpenAI → OpenAI | Peak RSS (MiB) | 250 | Valid | — |
| OpenAI → OpenAI | Peak RSS high-water mark (MiB) | 264 | Valid | — |
| OpenAI → OpenAI | Time to plateau (s) | — | Not measured | — |
| OpenAI → OpenAI | Memory growth (MiB/min) | 0.202 | Valid | — |
| OpenAI → OpenAI | Memory load duration (s) | 301 | Valid | — |
| OpenAI → OpenAI | Memory shape under load | 0 | Valid | — |
| OpenAI → OpenAI | Idle memory shape | -1 | Valid | — |
| OpenAI → OpenAI | Idle memory static | 0 | Valid | — |
| OpenAI → OpenAI | Idle memory growth (MiB/min) | -13.96 | Valid | — |
| OpenAI → OpenAI | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI → OpenAI | Throughput (req/s) | 831 | concurrency 1; n=4,992 | 5 ms |
| OpenAI → OpenAI | Throughput (req/s) | 835 | concurrency 2; n=5,012 | 10 ms |
| OpenAI → OpenAI | Throughput (req/s) | 836 | concurrency 8; n=5,018 | 50 ms |
| OpenAI → OpenAI | Throughput (req/s) | 836 | concurrency 8; n=5,018 | 100 ms |
| OpenAI → OpenAI | Throughput (req/s) | 836 | concurrency 8; n=5,018 | Unbounded |
| OpenAI Responses → OpenAI | — | — | Not declared: recorded note 1 | — |
| Anthropic → OpenAI | — | — | Not declared: recorded note 1 | — |
| Cohere → OpenAI | — | — | Not declared: recorded note 1 | — |
| Gemini → Cohere | — | — | Not declared: recorded note 1 | — |
| Cohere → Cohere | Added latency p50 (µs) | 1,128 | Valid · n=4,519 gateway / 248,527 direct | — |
| Cohere → Cohere | Added latency p99 (µs) | 3,714 | Valid · n=4,519 gateway / 248,527 direct | — |
| Cohere → Cohere | Gateway p99 at concurrency 1 (µs) | 3,750 | Valid · n=4,519 | — |
| Cohere → Cohere | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=248,527 | — |
| Cohere → Cohere | CPU per request (µs) | 1,331.2 | Valid | — |
| Cohere → Cohere | Requests per CPU-second | 751.2 | Valid | — |
| Cohere → Cohere | Cost window concurrency | 8 | Valid | — |
| Cohere → Cohere | Cost window core utilisation | 0.28 | Valid | — |
| Cohere → Cohere | Cost window successful requests | 5,063 | Valid | — |
| Cohere → Cohere | Cost window throughput (req/s) | 843 | Valid | — |
| Cohere → Cohere | Cost window threads | 22 | Valid | — |
| Cohere → Cohere | Non-voluntary context switches per request | 0.005 | Valid | — |
| Cohere → Cohere | Cost window major page faults | 0 | Valid | — |
| Cohere → Cohere | Added time to first event p50 (µs) | — | Untestable | — |
| Cohere → Cohere | Added time to first event p99 (µs) | — | Untestable | — |
| Cohere → Cohere | Added inter-frame gap p50 (µs) | — | Untestable | — |
| Cohere → Cohere | Added inter-frame gap p99 (µs) | — | Untestable | — |
| Cohere → Cohere | Streams sustained (frames/s) | — | Untestable | — |
| Cohere → Cohere | Streams sustained | — | Untestable | — |
| Cohere → Cohere | Idle RSS (MiB) | 125 | Valid | — |
| Cohere → Cohere | Steady-state RSS (MiB) | 245.9 | Valid | — |
| Cohere → Cohere | Recovered RSS (MiB) | 141.5 | Valid | — |
| Cohere → Cohere | Peak RSS (MiB) | 249.9 | Valid | — |
| Cohere → Cohere | Peak RSS high-water mark (MiB) | 264 | Valid | — |
| Cohere → Cohere | Time to plateau (s) | — | Not measured | — |
| Cohere → Cohere | Memory growth (MiB/min) | 0.806 | Valid | — |
| Cohere → Cohere | Memory load duration (s) | 301 | Valid | — |
| Cohere → Cohere | Memory shape under load | 0 | Valid | — |
| Cohere → Cohere | Idle memory shape | -1 | Valid | — |
| Cohere → Cohere | Idle memory static | 0 | Valid | — |
| Cohere → Cohere | Idle memory growth (MiB/min) | -14.17 | Valid | — |
| Cohere → Cohere | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| Cohere → Cohere | Throughput (req/s) | 860 | concurrency 2; n=5,161 | 5 ms |
| Cohere → Cohere | Throughput (req/s) | 860 | concurrency 2; n=5,161 | 10 ms |
| Cohere → Cohere | Throughput (req/s) | 860 | concurrency 2; n=5,161 | 50 ms |
| Cohere → Cohere | Throughput (req/s) | 860 | concurrency 2; n=5,161 | 100 ms |
| Cohere → Cohere | Throughput (req/s) | 860 | concurrency 2; n=5,161 | Unbounded |
| Bedrock → Cohere | — | — | Not declared: recorded note 1 | — |
| OpenAI → Cohere | Added latency p50 (µs) | 1,204 | Valid · n=4,255 gateway / 267,416 direct | — |
| OpenAI → Cohere | Added latency p99 (µs) | 3,792 | Valid · n=4,255 gateway / 267,416 direct | — |
| OpenAI → Cohere | Gateway p99 at concurrency 1 (µs) | 3,819 | Valid · n=4,255 | — |
| OpenAI → Cohere | Direct-to-mock p99 at concurrency 1 (µs) | 27 | Valid · n=267,416 | — |
| OpenAI → Cohere | CPU per request (µs) | 1,383.8 | Valid | — |
| OpenAI → Cohere | Requests per CPU-second | 722.6 | Valid | — |
| OpenAI → Cohere | Cost window concurrency | 8 | Valid | — |
| OpenAI → Cohere | Cost window core utilisation | 0.28 | Valid | — |
| OpenAI → Cohere | Cost window successful requests | 4,885 | Valid | — |
| OpenAI → Cohere | Cost window throughput (req/s) | 813 | Valid | — |
| OpenAI → Cohere | Cost window threads | 22 | Valid | — |
| OpenAI → Cohere | Non-voluntary context switches per request | 0.006 | Valid | — |
| OpenAI → Cohere | Cost window major page faults | 0 | Valid | — |
| OpenAI → Cohere | Added time to first event p50 (µs) | — | Untestable | — |
| OpenAI → Cohere | Added time to first event p99 (µs) | — | Untestable | — |
| OpenAI → Cohere | Added inter-frame gap p50 (µs) | — | Untestable | — |
| OpenAI → Cohere | Added inter-frame gap p99 (µs) | — | Untestable | — |
| OpenAI → Cohere | Streams sustained (frames/s) | — | Untestable | — |
| OpenAI → Cohere | Streams sustained | — | Untestable | — |
| OpenAI → Cohere | Idle RSS (MiB) | 124.9 | Valid | — |
| OpenAI → Cohere | Steady-state RSS (MiB) | 244.8 | Valid | — |
| OpenAI → Cohere | Recovered RSS (MiB) | 141.5 | Valid | — |
| OpenAI → Cohere | Peak RSS (MiB) | 250.2 | Valid | — |
| OpenAI → Cohere | Peak RSS high-water mark (MiB) | 264.8 | Valid | — |
| OpenAI → Cohere | Time to plateau (s) | — | Not measured | — |
| OpenAI → Cohere | Memory growth (MiB/min) | 0.676 | Valid | — |
| OpenAI → Cohere | Memory load duration (s) | 301 | Valid | — |
| OpenAI → Cohere | Memory shape under load | 0 | Valid | — |
| OpenAI → Cohere | Idle memory shape | -1 | Valid | — |
| OpenAI → Cohere | Idle memory static | 0 | Valid | — |
| OpenAI → Cohere | Idle memory growth (MiB/min) | -14.25 | Valid | — |
| OpenAI → Cohere | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI → Cohere | Throughput (req/s) | 821 | concurrency 2; n=4,929 | 5 ms |
| OpenAI → Cohere | Throughput (req/s) | 828 | concurrency 4; n=4,975 | 10 ms |
| OpenAI → Cohere | Throughput (req/s) | 828 | concurrency 4; n=4,975 | 50 ms |
| OpenAI → Cohere | Throughput (req/s) | 828 | concurrency 4; n=4,975 | 100 ms |
| OpenAI → Cohere | Throughput (req/s) | 828 | concurrency 4; n=4,975 | Unbounded |
| OpenAI Responses → Cohere | — | — | Not declared: recorded note 1 | — |
| Anthropic → Cohere | — | — | Not declared: recorded note 1 | — |
| Gemini → Anthropic | — | — | Not declared: recorded note 1 | — |
| Cohere → Anthropic | — | — | Not declared: recorded note 1 | — |
| Bedrock → Anthropic | — | — | Not declared: recorded note 1 | — |
| Anthropic → Anthropic | Added latency p50 (µs) | 1,196 | Valid · n=4,332 gateway / 243,420 direct | — |
| Anthropic → Anthropic | Added latency p99 (µs) | 3,745 | Valid · n=4,332 gateway / 243,420 direct | — |
| Anthropic → Anthropic | Gateway p99 at concurrency 1 (µs) | 3,781 | Valid · n=4,332 | — |
| Anthropic → Anthropic | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=243,420 | — |
| Anthropic → Anthropic | CPU per request (µs) | 1,549.2 | Valid | — |
| Anthropic → Anthropic | Requests per CPU-second | 645.5 | Valid | — |
| Anthropic → Anthropic | Cost window concurrency | 8 | Valid | — |
| Anthropic → Anthropic | Cost window core utilisation | 0.3 | Valid | — |
| Anthropic → Anthropic | Cost window successful requests | 4,667 | Valid | — |
| Anthropic → Anthropic | Cost window throughput (req/s) | 777 | Valid | — |
| Anthropic → Anthropic | Cost window threads | 22 | Valid | — |
| Anthropic → Anthropic | Non-voluntary context switches per request | 0.007 | Valid | — |
| Anthropic → Anthropic | Cost window major page faults | 0 | Valid | — |
| Anthropic → Anthropic | Added time to first event p50 (µs) | 28,106 | Valid · n=100 gateway / 100 direct | — |
| Anthropic → Anthropic | Added time to first event p99 (µs) | 29,379 | Valid · n=100 gateway / 100 direct | — |
| Anthropic → Anthropic | Added inter-frame gap p50 (µs) | 7 | Valid · n=63 gateway / 63 direct | — |
| Anthropic → Anthropic | Added inter-frame gap p99 (µs) | 400 | Valid · n=63 gateway / 63 direct | — |
| Anthropic → Anthropic | Streams sustained (frames/s) | 4,968.8 | Valid | — |
| Anthropic → Anthropic | Streams sustained | 1,087 | Valid | — |
| Anthropic → Anthropic | Idle RSS (MiB) | 125 | Valid | — |
| Anthropic → Anthropic | Steady-state RSS (MiB) | 245.3 | Valid | — |
| Anthropic → Anthropic | Recovered RSS (MiB) | 141.7 | Valid | — |
| Anthropic → Anthropic | Peak RSS (MiB) | 253.7 | Valid | — |
| Anthropic → Anthropic | Peak RSS high-water mark (MiB) | 268.4 | Valid | — |
| Anthropic → Anthropic | Time to plateau (s) | — | Not measured | — |
| Anthropic → Anthropic | Memory growth (MiB/min) | -1.46 | Valid | — |
| Anthropic → Anthropic | Memory load duration (s) | 301 | Valid | — |
| Anthropic → Anthropic | Memory shape under load | 0 | Valid | — |
| Anthropic → Anthropic | Idle memory shape | -1 | Valid | — |
| Anthropic → Anthropic | Idle memory static | 0 | Valid | — |
| Anthropic → Anthropic | Idle memory growth (MiB/min) | -14.09 | Valid | — |
| Anthropic → Anthropic | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| Anthropic → Anthropic | Throughput (req/s) | 823 | concurrency 2; n=4,941 | 5 ms |
| Anthropic → Anthropic | Throughput (req/s) | 823 | concurrency 2; n=4,941 | 10 ms |
| Anthropic → Anthropic | Throughput (req/s) | 823 | concurrency 2; n=4,941 | 50 ms |
| Anthropic → Anthropic | Throughput (req/s) | 823 | concurrency 2; n=4,941 | 100 ms |
| Anthropic → Anthropic | Throughput (req/s) | 823 | concurrency 2; n=4,941 | Unbounded |
| OpenAI → Anthropic | Added latency p50 (µs) | 1,238 | Valid · n=4,169 gateway / 245,970 direct | — |
| OpenAI → Anthropic | Added latency p99 (µs) | 3,814 | Valid · n=4,169 gateway / 245,970 direct | — |
| OpenAI → Anthropic | Gateway p99 at concurrency 1 (µs) | 3,850 | Valid · n=4,169 | — |
| OpenAI → Anthropic | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=245,970 | — |
| OpenAI → Anthropic | CPU per request (µs) | 1,583.7 | Valid | — |
| OpenAI → Anthropic | Requests per CPU-second | 631.4 | Valid | — |
| OpenAI → Anthropic | Cost window concurrency | 8 | Valid | — |
| OpenAI → Anthropic | Cost window core utilisation | 0.297 | Valid | — |
| OpenAI → Anthropic | Cost window successful requests | 4,521 | Valid | — |
| OpenAI → Anthropic | Cost window throughput (req/s) | 753 | Valid | — |
| OpenAI → Anthropic | Cost window threads | 22 | Valid | — |
| OpenAI → Anthropic | Non-voluntary context switches per request | 0.007 | Valid | — |
| OpenAI → Anthropic | Cost window major page faults | 0 | Valid | — |
| OpenAI → Anthropic | Added time to first event p50 (µs) | 28,321 | Valid · n=100 gateway / 100 direct | — |
| OpenAI → Anthropic | Added time to first event p99 (µs) | 29,941 | Valid · n=100 gateway / 100 direct | — |
| OpenAI → Anthropic | Added inter-frame gap p50 (µs) | 2 | Valid · n=63 gateway / 63 direct | — |
| OpenAI → Anthropic | Added inter-frame gap p99 (µs) | 363 | Valid · n=63 gateway / 63 direct | — |
| OpenAI → Anthropic | Streams sustained (frames/s) | 4,586.9 | Valid | — |
| OpenAI → Anthropic | Streams sustained | 1,055 | Valid | — |
| OpenAI → Anthropic | Idle RSS (MiB) | 125.2 | Valid | — |
| OpenAI → Anthropic | Steady-state RSS (MiB) | 246.1 | Valid | — |
| OpenAI → Anthropic | Recovered RSS (MiB) | 141.9 | Valid | — |
| OpenAI → Anthropic | Peak RSS (MiB) | 250.6 | Valid | — |
| OpenAI → Anthropic | Peak RSS high-water mark (MiB) | 265.1 | Valid | — |
| OpenAI → Anthropic | Time to plateau (s) | — | Not measured | — |
| OpenAI → Anthropic | Memory growth (MiB/min) | 1.69 | Valid | — |
| OpenAI → Anthropic | Memory load duration (s) | 301 | Valid | — |
| OpenAI → Anthropic | Memory shape under load | 0 | Valid | — |
| OpenAI → Anthropic | Idle memory shape | -1 | Valid | — |
| OpenAI → Anthropic | Idle memory static | 0 | Valid | — |
| OpenAI → Anthropic | Idle memory growth (MiB/min) | -13.91 | Valid | — |
| OpenAI → Anthropic | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI → Anthropic | Throughput (req/s) | 824 | concurrency 2; n=4,948 | 5 ms |
| OpenAI → Anthropic | Throughput (req/s) | 824 | concurrency 2; n=4,948 | 10 ms |
| OpenAI → Anthropic | Throughput (req/s) | 824 | concurrency 2; n=4,948 | 50 ms |
| OpenAI → Anthropic | Throughput (req/s) | 824 | concurrency 2; n=4,948 | 100 ms |
| OpenAI → Anthropic | Throughput (req/s) | 824 | concurrency 2; n=4,948 | Unbounded |
| OpenAI Responses → Anthropic | — | — | Not declared: recorded note 1 | — |
| Gemini → Bedrock | — | — | Not declared: recorded note 1 | — |
| Bedrock → Bedrock | Added latency p50 (µs) | 1,316 | Valid · n=3,896 gateway / 249,657 direct | — |
| Bedrock → Bedrock | Added latency p99 (µs) | 4,150 | Valid · n=3,896 gateway / 249,657 direct | — |
| Bedrock → Bedrock | Gateway p99 at concurrency 1 (µs) | 4,186 | Valid · n=3,896 | — |
| Bedrock → Bedrock | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=249,657 | — |
| Bedrock → Bedrock | CPU per request (µs) | 1,534.8 | Valid | — |
| Bedrock → Bedrock | Requests per CPU-second | 651.5 | Valid | — |
| Bedrock → Bedrock | Cost window concurrency | 8 | Valid | — |
| Bedrock → Bedrock | Cost window core utilisation | 0.283 | Valid | — |
| Bedrock → Bedrock | Cost window successful requests | 4,450 | Valid | — |
| Bedrock → Bedrock | Cost window throughput (req/s) | 741 | Valid | — |
| Bedrock → Bedrock | Cost window threads | 22 | Valid | — |
| Bedrock → Bedrock | Non-voluntary context switches per request | 0.007 | Valid | — |
| Bedrock → Bedrock | Cost window major page faults | 0 | Valid | — |
| Bedrock → Bedrock | Added time to first event p50 (µs) | — | Untestable | — |
| Bedrock → Bedrock | Added time to first event p99 (µs) | — | Untestable | — |
| Bedrock → Bedrock | Added inter-frame gap p50 (µs) | — | Untestable | — |
| Bedrock → Bedrock | Added inter-frame gap p99 (µs) | — | Untestable | — |
| Bedrock → Bedrock | Streams sustained (frames/s) | — | Untestable | — |
| Bedrock → Bedrock | Streams sustained | — | Untestable | — |
| Bedrock → Bedrock | Idle RSS (MiB) | 125 | Valid | — |
| Bedrock → Bedrock | Steady-state RSS (MiB) | 248.5 | Valid | — |
| Bedrock → Bedrock | Recovered RSS (MiB) | 142.3 | Valid | — |
| Bedrock → Bedrock | Peak RSS (MiB) | 253.1 | Valid | — |
| Bedrock → Bedrock | Peak RSS high-water mark (MiB) | 268.1 | Valid | — |
| Bedrock → Bedrock | Time to plateau (s) | — | Not measured | — |
| Bedrock → Bedrock | Memory growth (MiB/min) | 0.327 | Valid | — |
| Bedrock → Bedrock | Memory load duration (s) | 301 | Valid | — |
| Bedrock → Bedrock | Memory shape under load | 0 | Valid | — |
| Bedrock → Bedrock | Idle memory shape | -1 | Valid | — |
| Bedrock → Bedrock | Idle memory static | 0 | Valid | — |
| Bedrock → Bedrock | Idle memory growth (MiB/min) | -14.19 | Valid | — |
| Bedrock → Bedrock | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| Bedrock → Bedrock | Throughput (req/s) | 757 | concurrency 2; n=4,547 | 5 ms |
| Bedrock → Bedrock | Throughput (req/s) | 757 | concurrency 2; n=4,547 | 10 ms |
| Bedrock → Bedrock | Throughput (req/s) | 757 | concurrency 2; n=4,547 | 50 ms |
| Bedrock → Bedrock | Throughput (req/s) | 757 | concurrency 2; n=4,547 | 100 ms |
| Bedrock → Bedrock | Throughput (req/s) | 757 | concurrency 2; n=4,547 | Unbounded |
| OpenAI Responses → Bedrock | — | — | Not declared: recorded note 1 | — |
| Anthropic → Bedrock | Added latency p50 (µs) | 1,438 | Valid · n=3,579 gateway / 262,193 direct | — |
| Anthropic → Bedrock | Added latency p99 (µs) | 4,467 | Valid · n=3,579 gateway / 262,193 direct | — |
| Anthropic → Bedrock | Gateway p99 at concurrency 1 (µs) | 4,494 | Valid · n=3,579 | — |
| Anthropic → Bedrock | Direct-to-mock p99 at concurrency 1 (µs) | 27 | Valid · n=262,193 | — |
| Anthropic → Bedrock | CPU per request (µs) | 1,632.6 | Valid | — |
| Anthropic → Bedrock | Requests per CPU-second | 612.5 | Valid | — |
| Anthropic → Bedrock | Cost window concurrency | 8 | Valid | — |
| Anthropic → Bedrock | Cost window core utilisation | 0.282 | Valid | — |
| Anthropic → Bedrock | Cost window successful requests | 4,159 | Valid | — |
| Anthropic → Bedrock | Cost window throughput (req/s) | 693 | Valid | — |
| Anthropic → Bedrock | Cost window threads | 22 | Valid | — |
| Anthropic → Bedrock | Non-voluntary context switches per request | 0.007 | Valid | — |
| Anthropic → Bedrock | Cost window major page faults | 0 | Valid | — |
| Anthropic → Bedrock | Added time to first event p50 (µs) | — | Untestable | — |
| Anthropic → Bedrock | Added time to first event p99 (µs) | — | Untestable | — |
| Anthropic → Bedrock | Added inter-frame gap p50 (µs) | — | Untestable | — |
| Anthropic → Bedrock | Added inter-frame gap p99 (µs) | — | Untestable | — |
| Anthropic → Bedrock | Streams sustained (frames/s) | — | Untestable | — |
| Anthropic → Bedrock | Streams sustained | — | Untestable | — |
| Anthropic → Bedrock | Idle RSS (MiB) | 125.1 | Valid | — |
| Anthropic → Bedrock | Steady-state RSS (MiB) | 250.3 | Valid | — |
| Anthropic → Bedrock | Recovered RSS (MiB) | 142.5 | Valid | — |
| Anthropic → Bedrock | Peak RSS (MiB) | 254.4 | Valid | — |
| Anthropic → Bedrock | Peak RSS high-water mark (MiB) | 268.3 | Valid | — |
| Anthropic → Bedrock | Time to plateau (s) | — | Not measured | — |
| Anthropic → Bedrock | Memory growth (MiB/min) | -2.66 | Valid | — |
| Anthropic → Bedrock | Memory load duration (s) | 301 | Valid | — |
| Anthropic → Bedrock | Memory shape under load | 0 | Valid | — |
| Anthropic → Bedrock | Idle memory shape | -1 | Valid | — |
| Anthropic → Bedrock | Idle memory static | 0 | Valid | — |
| Anthropic → Bedrock | Idle memory growth (MiB/min) | -13.93 | Valid | — |
| Anthropic → Bedrock | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| Anthropic → Bedrock | Throughput (req/s) | 703 | concurrency 1; n=4,223 | 5 ms |
| Anthropic → Bedrock | Throughput (req/s) | 706 | concurrency 2; n=4,237 | 10 ms |
| Anthropic → Bedrock | Throughput (req/s) | 706 | concurrency 2; n=4,237 | 50 ms |
| Anthropic → Bedrock | Throughput (req/s) | 706 | concurrency 2; n=4,237 | 100 ms |
| Anthropic → Bedrock | Throughput (req/s) | 706 | concurrency 2; n=4,237 | Unbounded |
| Cohere → Bedrock | — | — | Not declared: recorded note 1 | — |
| OpenAI → Bedrock | Added latency p50 (µs) | 1,430 | Valid · n=3,563 gateway / 269,629 direct | — |
| OpenAI → Bedrock | Added latency p99 (µs) | 4,577 | Valid · n=3,563 gateway / 269,629 direct | — |
| OpenAI → Bedrock | Gateway p99 at concurrency 1 (µs) | 4,604 | Valid · n=3,563 | — |
| OpenAI → Bedrock | Direct-to-mock p99 at concurrency 1 (µs) | 27 | Valid · n=269,629 | — |
| OpenAI → Bedrock | CPU per request (µs) | 1,664.2 | Valid | — |
| OpenAI → Bedrock | Requests per CPU-second | 600.9 | Valid | — |
| OpenAI → Bedrock | Cost window concurrency | 8 | Valid | — |
| OpenAI → Bedrock | Cost window core utilisation | 0.283 | Valid | — |
| OpenAI → Bedrock | Cost window successful requests | 4,098 | Valid | — |
| OpenAI → Bedrock | Cost window throughput (req/s) | 682 | Valid | — |
| OpenAI → Bedrock | Cost window threads | 22 | Valid | — |
| OpenAI → Bedrock | Non-voluntary context switches per request | 0.008 | Valid | — |
| OpenAI → Bedrock | Cost window major page faults | 0 | Valid | — |
| OpenAI → Bedrock | Added time to first event p50 (µs) | — | Untestable | — |
| OpenAI → Bedrock | Added time to first event p99 (µs) | — | Untestable | — |
| OpenAI → Bedrock | Added inter-frame gap p50 (µs) | — | Untestable | — |
| OpenAI → Bedrock | Added inter-frame gap p99 (µs) | — | Untestable | — |
| OpenAI → Bedrock | Streams sustained (frames/s) | — | Untestable | — |
| OpenAI → Bedrock | Streams sustained | — | Untestable | — |
| OpenAI → Bedrock | Idle RSS (MiB) | 125.2 | Valid | — |
| OpenAI → Bedrock | Steady-state RSS (MiB) | 248.4 | Valid | — |
| OpenAI → Bedrock | Recovered RSS (MiB) | 142.8 | Valid | — |
| OpenAI → Bedrock | Peak RSS (MiB) | 253.9 | Valid | — |
| OpenAI → Bedrock | Peak RSS high-water mark (MiB) | 268.4 | Valid | — |
| OpenAI → Bedrock | Time to plateau (s) | — | Not measured | — |
| OpenAI → Bedrock | Memory growth (MiB/min) | -1.45 | Valid | — |
| OpenAI → Bedrock | Memory load duration (s) | 301 | Valid | — |
| OpenAI → Bedrock | Memory shape under load | 0 | Valid | — |
| OpenAI → Bedrock | Idle memory shape | -1 | Valid | — |
| OpenAI → Bedrock | Idle memory static | 0 | Valid | — |
| OpenAI → Bedrock | Idle memory growth (MiB/min) | -14.02 | Valid | — |
| OpenAI → Bedrock | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI → Bedrock | Throughput (req/s) | 685 | concurrency 1; n=4,114 | 5 ms |
| OpenAI → Bedrock | Throughput (req/s) | 696 | concurrency 2; n=4,177 | 10 ms |
| OpenAI → Bedrock | Throughput (req/s) | 696 | concurrency 2; n=4,177 | 50 ms |
| OpenAI → Bedrock | Throughput (req/s) | 696 | concurrency 2; n=4,177 | 100 ms |
| OpenAI → Bedrock | Throughput (req/s) | 696 | concurrency 2; n=4,177 | Unbounded |
| OpenAI Responses → Gemini | — | — | Not declared: recorded note 1 | — |
| Anthropic → Gemini | — | — | Not declared: recorded note 1 | — |
| Gemini → Gemini | Added latency p50 (µs) | 1,145 | Valid · n=4,389 gateway / 249,981 direct | — |
| Gemini → Gemini | Added latency p99 (µs) | 4,018 | Valid · n=4,389 gateway / 249,981 direct | — |
| Gemini → Gemini | Gateway p99 at concurrency 1 (µs) | 4,054 | Valid · n=4,389 | — |
| Gemini → Gemini | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=249,981 | — |
| Gemini → Gemini | CPU per request (µs) | 1,361.8 | Valid | — |
| Gemini → Gemini | Requests per CPU-second | 734.3 | Valid | — |
| Gemini → Gemini | Cost window concurrency | 8 | Valid | — |
| Gemini → Gemini | Cost window core utilisation | 0.286 | Valid | — |
| Gemini → Gemini | Cost window successful requests | 5,052 | Valid | — |
| Gemini → Gemini | Cost window throughput (req/s) | 841 | Valid | — |
| Gemini → Gemini | Cost window threads | 22 | Valid | — |
| Gemini → Gemini | Non-voluntary context switches per request | 0.008 | Valid | — |
| Gemini → Gemini | Cost window major page faults | 0 | Valid | — |
| Gemini → Gemini | Added time to first event p50 (µs) | — | Untestable | — |
| Gemini → Gemini | Added time to first event p99 (µs) | — | Untestable | — |
| Gemini → Gemini | Added inter-frame gap p50 (µs) | — | Untestable | — |
| Gemini → Gemini | Added inter-frame gap p99 (µs) | — | Untestable | — |
| Gemini → Gemini | Streams sustained (frames/s) | — | Untestable | — |
| Gemini → Gemini | Streams sustained | — | Untestable | — |
| Gemini → Gemini | Idle RSS (MiB) | 124.1 | Valid | — |
| Gemini → Gemini | Steady-state RSS (MiB) | 245.7 | Valid | — |
| Gemini → Gemini | Recovered RSS (MiB) | 140 | Valid | — |
| Gemini → Gemini | Peak RSS (MiB) | 249.9 | Valid | — |
| Gemini → Gemini | Peak RSS high-water mark (MiB) | 263.9 | Valid | — |
| Gemini → Gemini | Time to plateau (s) | — | Not measured | — |
| Gemini → Gemini | Memory growth (MiB/min) | -0.813 | Valid | — |
| Gemini → Gemini | Memory load duration (s) | 301 | Valid | — |
| Gemini → Gemini | Memory shape under load | 0 | Valid | — |
| Gemini → Gemini | Idle memory shape | -1 | Valid | — |
| Gemini → Gemini | Idle memory static | 0 | Valid | — |
| Gemini → Gemini | Idle memory growth (MiB/min) | -14.01 | Valid | — |
| Gemini → Gemini | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| Gemini → Gemini | Throughput (req/s) | 865 | concurrency 2; n=5,193 | 5 ms |
| Gemini → Gemini | Throughput (req/s) | 868 | concurrency 4; n=5,212 | 10 ms |
| Gemini → Gemini | Throughput (req/s) | 868 | concurrency 4; n=5,212 | 50 ms |
| Gemini → Gemini | Throughput (req/s) | 868 | concurrency 4; n=5,212 | 100 ms |
| Gemini → Gemini | Throughput (req/s) | 868 | concurrency 4; n=5,212 | Unbounded |
| OpenAI → Gemini | Added latency p50 (µs) | 1,199 | Valid · n=4,241 gateway / 249,376 direct | — |
| OpenAI → Gemini | Added latency p99 (µs) | 3,969 | Valid · n=4,241 gateway / 249,376 direct | — |
| OpenAI → Gemini | Gateway p99 at concurrency 1 (µs) | 4,005 | Valid · n=4,241 | — |
| OpenAI → Gemini | Direct-to-mock p99 at concurrency 1 (µs) | 36 | Valid · n=249,376 | — |
| OpenAI → Gemini | CPU per request (µs) | 1,420.5 | Valid | — |
| OpenAI → Gemini | Requests per CPU-second | 704 | Valid | — |
| OpenAI → Gemini | Cost window concurrency | 8 | Valid | — |
| OpenAI → Gemini | Cost window core utilisation | 0.283 | Valid | — |
| OpenAI → Gemini | Cost window successful requests | 4,801 | Valid | — |
| OpenAI → Gemini | Cost window throughput (req/s) | 800 | Valid | — |
| OpenAI → Gemini | Cost window threads | 22 | Valid | — |
| OpenAI → Gemini | Non-voluntary context switches per request | 0.006 | Valid | — |
| OpenAI → Gemini | Cost window major page faults | 0 | Valid | — |
| OpenAI → Gemini | Added time to first event p50 (µs) | — | Untestable | — |
| OpenAI → Gemini | Added time to first event p99 (µs) | — | Untestable | — |
| OpenAI → Gemini | Added inter-frame gap p50 (µs) | — | Untestable | — |
| OpenAI → Gemini | Added inter-frame gap p99 (µs) | — | Untestable | — |
| OpenAI → Gemini | Streams sustained (frames/s) | — | Untestable | — |
| OpenAI → Gemini | Streams sustained | — | Untestable | — |
| OpenAI → Gemini | Idle RSS (MiB) | 125.2 | Valid | — |
| OpenAI → Gemini | Steady-state RSS (MiB) | 247 | Valid | — |
| OpenAI → Gemini | Recovered RSS (MiB) | 142.2 | Valid | — |
| OpenAI → Gemini | Peak RSS (MiB) | 250.6 | Valid | — |
| OpenAI → Gemini | Peak RSS high-water mark (MiB) | 265.4 | Valid | — |
| OpenAI → Gemini | Time to plateau (s) | — | Not measured | — |
| OpenAI → Gemini | Memory growth (MiB/min) | -1.36 | Valid | — |
| OpenAI → Gemini | Memory load duration (s) | 301 | Valid | — |
| OpenAI → Gemini | Memory shape under load | 0 | Valid | — |
| OpenAI → Gemini | Idle memory shape | -1 | Valid | — |
| OpenAI → Gemini | Idle memory static | 0 | Valid | — |
| OpenAI → Gemini | Idle memory growth (MiB/min) | -14.12 | Valid | — |
| OpenAI → Gemini | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI → Gemini | Throughput (req/s) | 800 | concurrency 1; n=4,805 | 5 ms |
| OpenAI → Gemini | Throughput (req/s) | 809 | concurrency 2; n=4,857 | 10 ms |
| OpenAI → Gemini | Throughput (req/s) | 809 | concurrency 2; n=4,857 | 50 ms |
| OpenAI → Gemini | Throughput (req/s) | 809 | concurrency 2; n=4,857 | 100 ms |
| OpenAI → Gemini | Throughput (req/s) | 809 | concurrency 2; n=4,857 | Unbounded |
| Bedrock → Gemini | — | — | Not declared: recorded note 1 | — |
| Cohere → Gemini | — | — | Not declared: recorded note 1 | — |
| Anthropic → OpenAI Responses | — | — | Not declared: recorded note 1 | — |
| OpenAI Responses → OpenAI Responses | Added latency p50 (µs) | 1,238 | Valid · n=4,169 gateway / 265,251 direct | — |
| OpenAI Responses → OpenAI Responses | Added latency p99 (µs) | 3,854 | Valid · n=4,169 gateway / 265,251 direct | — |
| OpenAI Responses → OpenAI Responses | Gateway p99 at concurrency 1 (µs) | 3,881 | Valid · n=4,169 | — |
| OpenAI Responses → OpenAI Responses | Direct-to-mock p99 at concurrency 1 (µs) | 27 | Valid · n=265,251 | — |
| OpenAI Responses → OpenAI Responses | CPU per request (µs) | 1,427.7 | Valid | — |
| OpenAI Responses → OpenAI Responses | Requests per CPU-second | 700.4 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window concurrency | 8 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window core utilisation | 0.285 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window successful requests | 4,805 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window throughput (req/s) | 800 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window threads | 22 | Valid | — |
| OpenAI Responses → OpenAI Responses | Non-voluntary context switches per request | 0.004 | Valid | — |
| OpenAI Responses → OpenAI Responses | Cost window major page faults | 0 | Valid | — |
| OpenAI Responses → OpenAI Responses | Added time to first event p50 (µs) | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Added time to first event p99 (µs) | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Added inter-frame gap p50 (µs) | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Added inter-frame gap p99 (µs) | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Streams sustained (frames/s) | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Streams sustained | — | Untestable | — |
| OpenAI Responses → OpenAI Responses | Idle RSS (MiB) | 125 | Valid | — |
| OpenAI Responses → OpenAI Responses | Steady-state RSS (MiB) | 243.7 | Valid | — |
| OpenAI Responses → OpenAI Responses | Recovered RSS (MiB) | 142 | Valid | — |
| OpenAI Responses → OpenAI Responses | Peak RSS (MiB) | 249.5 | Valid | — |
| OpenAI Responses → OpenAI Responses | Peak RSS high-water mark (MiB) | 264.3 | Valid | — |
| OpenAI Responses → OpenAI Responses | Time to plateau (s) | — | Not measured | — |
| OpenAI Responses → OpenAI Responses | Memory growth (MiB/min) | -1.94 | Valid | — |
| OpenAI Responses → OpenAI Responses | Memory load duration (s) | 301 | Valid | — |
| OpenAI Responses → OpenAI Responses | Memory shape under load | 0 | Valid | — |
| OpenAI Responses → OpenAI Responses | Idle memory shape | -1 | Valid | — |
| OpenAI Responses → OpenAI Responses | Idle memory static | 0 | Valid | — |
| OpenAI Responses → OpenAI Responses | Idle memory growth (MiB/min) | -14.08 | Valid | — |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | — | Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound | 1 ms |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | 809 | concurrency 2; n=4,855 | 5 ms |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | 809 | concurrency 2; n=4,855 | 10 ms |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | 809 | concurrency 2; n=4,855 | 50 ms |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | 809 | concurrency 2; n=4,855 | 100 ms |
| OpenAI Responses → OpenAI Responses | Throughput (req/s) | 809 | concurrency 2; n=4,855 | Unbounded |
| OpenAI → OpenAI Responses | — | — | Not declared: recorded note 1 | — |
| Gemini → OpenAI Responses | — | — | Not declared: recorded note 1 | — |
| Bedrock → OpenAI Responses | — | — | Not declared: recorded note 1 | — |
| Cohere → OpenAI Responses | — | — | Not declared: recorded note 1 | — |
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 → OpenAI | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → OpenAI | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| OpenAI → OpenAI | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → OpenAI | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → OpenAI | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Cohere → OpenAI | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → Cohere | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Cohere → Cohere | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| Bedrock → Cohere | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| OpenAI → Cohere | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → Cohere | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → Cohere | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → Anthropic | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Cohere → Anthropic | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Bedrock → Anthropic | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → Anthropic | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI → Anthropic | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → Anthropic | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → Bedrock | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Bedrock → Bedrock | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → Bedrock | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → Bedrock | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| Cohere → Bedrock | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| OpenAI → Bedrock | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → Gemini | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → Gemini | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → Gemini | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI → Gemini | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| Bedrock → Gemini | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Cohere → Gemini | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Anthropic → OpenAI Responses | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| OpenAI Responses → OpenAI Responses | Served | — | — | — | Recorded run evidence | — | portkeyai/gateway:1.15.2 |
| OpenAI → OpenAI Responses | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Gemini → OpenAI Responses | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Bedrock → OpenAI Responses | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
| Cohere → OpenAI Responses | Not declared | — | recorded note 1 | recorded note 1 | recorded note 2 | — | portkeyai/gateway:1.15.2 |
- Harness commit
d240badd966933fa19eec0606b2d1f3945f18ad2- Mock SHA-256
99d01f709d8d9c1f49055aa265d5eb7475264d6b513911fe317aee74a71190f8- Load generator SHA-256
not recorded- net.core.somaxconn
- 65,535
- net.ipv4.tcp_max_syn_backlog
- 65,535
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 | HTTP 200: 1 | 0 |
| OpenAI → OpenAI | 2 | HTTP 200: 2 | 0 |
| OpenAI → OpenAI | 4 | HTTP 200: 4 | 0 |
| OpenAI → OpenAI | 8 | HTTP 200: 8 | 0 |
| OpenAI → OpenAI | 16 | HTTP 200: 16 | 0 |
| OpenAI → OpenAI | 32 | HTTP 200: 32 | 0 |
| OpenAI → OpenAI | 64 | HTTP 200: 64 | 0 |
| OpenAI → OpenAI | 128 | HTTP 200: 128 | 0 |
| OpenAI → OpenAI | 256 | HTTP 200: 256 | 0 |
| OpenAI → OpenAI | 512 | HTTP 200: 512 | 0 |
| OpenAI → OpenAI | 1,024 | HTTP 200: 1,024 | 0 |
| OpenAI → OpenAI | 2,048 | HTTP 200: 1,810 | 238 |
| OpenAI → OpenAI | 1,536 | HTTP 200: 703 | 833 |
| OpenAI → OpenAI | 1,280 | HTTP 200: 1,280 | 0 |
| OpenAI → OpenAI | 1,408 | HTTP 200: 1,408 | 0 |
| OpenAI → OpenAI | 1,472 | HTTP 200: 1,472 | 0 |
| OpenAI → OpenAI | 1,504 | HTTP 200: 1,504 | 0 |
| OpenAI → OpenAI | 1,520 | HTTP 200: 1,507 | 13 |
| OpenAI → OpenAI | 1,512 | HTTP 200: 1,512 | 0 |
| OpenAI → OpenAI | 1,516 | HTTP 200: 1,516 | 0 |
| OpenAI → OpenAI | 1,518 | HTTP 200: 1,518 | 0 |
| OpenAI → OpenAI | 1,519 | HTTP 200: 1,519 | 0 |
| OpenAI → OpenAI | 1,519 | HTTP 200: 1,445 | 74 |
| OpenAI → OpenAI | 1,519 | HTTP 200: 1,519 | 0 |
| Anthropic → Anthropic | 1 | HTTP 200: 1 | 0 |
| Anthropic → Anthropic | 2 | HTTP 200: 2 | 0 |
| Anthropic → Anthropic | 4 | HTTP 200: 4 | 0 |
| Anthropic → Anthropic | 8 | HTTP 200: 8 | 0 |
| Anthropic → Anthropic | 16 | HTTP 200: 16 | 0 |
| Anthropic → Anthropic | 32 | HTTP 200: 32 | 0 |
| Anthropic → Anthropic | 64 | HTTP 200: 64 | 0 |
| Anthropic → Anthropic | 128 | HTTP 200: 128 | 0 |
| Anthropic → Anthropic | 256 | HTTP 200: 256 | 0 |
| Anthropic → Anthropic | 512 | HTTP 200: 512 | 0 |
| Anthropic → Anthropic | 1,024 | HTTP 200: 1,024 | 0 |
| Anthropic → Anthropic | 2,048 | HTTP 200: 1,798 | 250 |
| Anthropic → Anthropic | 1,536 | HTTP 200: 1,151 | 385 |
| Anthropic → Anthropic | 1,280 | HTTP 200: 918 | 362 |
| Anthropic → Anthropic | 1,152 | HTTP 200: 919 | 233 |
| Anthropic → Anthropic | 1,088 | HTTP 200: 1,088 | 0 |
| Anthropic → Anthropic | 1,056 | HTTP 200: 1,056 | 0 |
| Anthropic → Anthropic | 1,072 | HTTP 200: 1,072 | 0 |
| Anthropic → Anthropic | 1,080 | HTTP 200: 1,080 | 0 |
| Anthropic → Anthropic | 1,084 | HTTP 200: 1,084 | 0 |
| Anthropic → Anthropic | 1,086 | HTTP 200: 1,086 | 0 |
| Anthropic → Anthropic | 1,087 | HTTP 200: 1,087 | 0 |
| Anthropic → Anthropic | 1,087 | HTTP 200: 1,087 | 0 |
| Anthropic → Anthropic | 1,087 | HTTP 200: 1,087 | 0 |
| OpenAI → Anthropic | 1 | HTTP 200: 1 | 0 |
| OpenAI → Anthropic | 2 | HTTP 200: 2 | 0 |
| OpenAI → Anthropic | 4 | HTTP 200: 4 | 0 |
| OpenAI → Anthropic | 8 | HTTP 200: 8 | 0 |
| OpenAI → Anthropic | 16 | HTTP 200: 16 | 0 |
| OpenAI → Anthropic | 32 | HTTP 200: 32 | 0 |
| OpenAI → Anthropic | 64 | HTTP 200: 64 | 0 |
| OpenAI → Anthropic | 128 | HTTP 200: 128 | 0 |
| OpenAI → Anthropic | 256 | HTTP 200: 256 | 0 |
| OpenAI → Anthropic | 512 | HTTP 200: 512 | 0 |
| OpenAI → Anthropic | 1,024 | HTTP 200: 1,024 | 0 |
| OpenAI → Anthropic | 2,048 | HTTP 200: 1,626 | 422 |
| OpenAI → Anthropic | 1,536 | HTTP 200: 1,211 | 325 |
| OpenAI → Anthropic | 1,280 | HTTP 200: 854 | 426 |
| OpenAI → Anthropic | 1,152 | HTTP 200: 950 | 202 |
| OpenAI → Anthropic | 1,088 | HTTP 200: 724 | 364 |
| OpenAI → Anthropic | 1,056 | HTTP 200: 1,050 | 6 |
| OpenAI → Anthropic | 1,040 | HTTP 200: 1,040 | 0 |
| OpenAI → Anthropic | 1,048 | HTTP 200: 1,048 | 0 |
| OpenAI → Anthropic | 1,052 | HTTP 200: 1,052 | 0 |
| OpenAI → Anthropic | 1,054 | HTTP 200: 1,054 | 0 |
| OpenAI → Anthropic | 1,055 | HTTP 200: 1,055 | 0 |
| OpenAI → Anthropic | 1,055 | HTTP 200: 1,055 | 0 |
| OpenAI → Anthropic | 1,055 | HTTP 200: 1,055 | 0 |
Recorded notes
Text recorded with the cells above. Each note is shown once; rows refer to it by number.
- Pinned v1.15.2 source ca7712947ec8c0887803990b746740d36011c697: src/index.ts registers chat-completions, messages, responses and raw proxy routes; provider operation tables determine translations. Recorder preflight proves 11 exact protocol pairs. Raw proxy passes native requests through without cross-protocol conversion, so only its matching diagonal is enabled. OpenAI Chat Completions and Responses share a provider but remain distinct upstream operations. Evidence: results/diagnostics/refresh-configuration/portkey-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-portkey-670dce412ba19689b30f84f64cb8b9c31fc07125970bd210617247c56c23bfcc · complete raw record including RSS series
Configuration recorded by the run
Not recorded.
Compare a shared protocol cell
- Portkey vs agentgateway
- Portkey vs AISIX (api7)
- Portkey vs APISIX
- Portkey vs Bifrost
- Portkey vs Busbar
- Portkey vs GoModel
- Portkey vs Helicone
- Portkey vs Higress
- Portkey vs Kong
- Portkey vs LiteLLM · Python
- Portkey vs LiteLLM · Rust
- Portkey vs One-API
- Portkey vs Plano
- Portkey vs TensorZero