OR
OpenRouter
OpenRouter Inc. · aggregator · est. 2023
Universal LLM gateway. 200+ models, one API key, transparent pricing.
aggregator open-weight frontier
Attested runs
1
on this provider
Models
10
served
Benchmarks
15
covered
Avg drift
0.0pp
vs canonical (n=0)
Default quant
FP-BF16
precision
Attested models on this provider
| Model | Benchmark | OpenRouter | Canonical | Drift |
|---|---|---|---|---|
| cohere/command-r-plus Self-reported | humaneval | 76.4 | — | — |
| cohere/command-r-plus Self-reported | mmlu-pro | 62.0 | — | — |
| deepseek/deepseek-v4 Self-reported | gpqa | 68.7 | — | — |
| deepseek/deepseek-v4 Self-reported | gsm8k | 95.0 | — | — |
| deepseek/deepseek-v4 Self-reported | humaneval | 92.2 | — | — |
| deepseek/deepseek-v4 Self-reported | livecodebench | 58.6 | — | — |
| deepseek/deepseek-v4 Self-reported | math | 81.0 | — | — |
| deepseek/deepseek-v4 Self-reported | mbpp | 89.8 | — | — |
| deepseek/deepseek-v4 Self-reported | swe-bench-verified | 56.4 | — | — |
| deepseek/deepseek-v4 Self-reported | tau-bench | 41.0 | — | — |
| google/gemini-2.6-flash Self-reported | gpqa | 58.0 | — | — |
| google/gemini-2.6-flash Self-reported | humaneval | 88.4 | — | — |
| google/gemini-2.6-flash Self-reported | mbpp | 85.0 | — | — |
| google/gemini-2.6-flash Self-reported | mmlu-pro | 70.2 | — | — |
| google/gemini-2.6-pro Self-reported | arc-agi | 19.4 | — | — |
| google/gemini-2.6-pro Self-reported | gaia | 36.5 | — | — |
| google/gemini-2.6-pro Self-reported | gpqa | 72.4 | — | — |
| google/gemini-2.6-pro Self-reported | gsm8k | 96.1 | — | — |
| google/gemini-2.6-pro Self-reported | hellaswag | 94.7 | — | — |
| google/gemini-2.6-pro Self-reported | humaneval | 94.0 | — | — |
| google/gemini-2.6-pro Self-reported | livecodebench | 55.2 | — | — |
| google/gemini-2.6-pro Self-reported | math | 82.1 | — | — |
| google/gemini-2.6-pro Self-reported | mbpp | 91.8 | — | — |
| google/gemini-2.6-pro Self-reported | mmlu | 91.0 | — | — |
| google/gemini-2.6-pro Self-reported | mmlu-pro | 84.8 | — | — |
| google/gemini-2.6-pro Self-reported | ruler | 92.3 | — | — |
| google/gemini-2.6-pro Self-reported | swe-bench-verified | 60.5 | — | — |
| google/gemini-2.6-pro Self-reported | truthfulqa | 64.8 | — | — |
| meta-llama/llama-3.3-70b Attested | mbpp | 76.4 ±18 | — | — |
| meta-llama/llama-4-405b Self-reported | gsm8k | 93.0 | — | — |
| meta-llama/llama-4-405b Self-reported | hellaswag | 92.6 | — | — |
| meta-llama/llama-4-405b Self-reported | humaneval | 85.2 | — | — |
| meta-llama/llama-4-405b Self-reported | mbpp | 82.0 | — | — |
| meta-llama/llama-4-405b Self-reported | mmlu | 86.8 | — | — |
| meta-llama/llama-4-405b Self-reported | mmlu-pro | 71.8 | — | — |
| mistral/large-3 Self-reported | gpqa | 54.8 | — | — |
| mistral/large-3 Self-reported | humaneval | 88.6 | — | — |
| mistral/large-3 Self-reported | mbpp | 86.2 | — | — |
| mistral/large-3 Self-reported | mmlu-pro | 76.4 | — | — |
| moonshotai/kimi-k2.6 Self-reported | humaneval | 90.4 | — | — |
| moonshotai/kimi-k2.6 Self-reported | math | 73.0 | — | — |
| moonshotai/kimi-k2.6 Self-reported | mbpp | 87.6 | — | — |
| moonshotai/kimi-k2.6 Self-reported | swe-bench-verified | 47.8 | — | — |
| qwen/qwen3.6-72b-instruct Self-reported | gpqa | 57.2 | — | — |
| qwen/qwen3.6-72b-instruct Self-reported | gsm8k | 94.2 | — | — |
| qwen/qwen3.6-72b-instruct Self-reported | humaneval | 87.9 | — | — |
| qwen/qwen3.6-72b-instruct Self-reported | math | 74.2 | — | — |
| qwen/qwen3.6-72b-instruct Self-reported | mbpp | 85.4 | — | — |
| x-ai/grok-4 Self-reported | gpqa | 64.5 | — | — |
| x-ai/grok-4 Self-reported | humaneval | 88.0 | — | — |
| x-ai/grok-4 Self-reported | math | 75.0 | — | — |
Provider Verified · $499/mo
Are you OpenRouter? Subscribe and own this page.
Unlimited multi-model attestations across your hosted catalog, drift alerts to Slack/webhook, customer-facing badge widget, and this openrouter page populated daily. See the full pitch →