295.2tok/s
Mean per test case, 274 tok/s over the run
4Ktokens
Answer plus reasoning
3.5Ktokens
87% of the output
1.5M
851.3K in, 616.7K out
43m 58s
17.2s per caseRequest time, summed
$0.83
$0.0054 per caseAt the provider's list price
Every test case
One square per test case. Select a square to read the answer.
55Agents8 of 21 passed
81Structured Extraction17 of 21 passed
48Coding10 of 21 passed
69Grounded Answers15 of 21 passed
78Instruction Following14 of 21 passed
57Business Logic12 of 21 passed
51Security & permissions8 of 21 passed
17Diagnosis & investigation1 of 6 passed
Speed and output per test case
One bar per test case. Hover to inspect a case, select it to read the answer.
Peak 540 tok/s
ReasoningAnswer
Test by test
Every answer with its reasoning and the checks it had to pass.
Reproduce this run
The prompts stay private so models can't train on them. The hashes pin the exact suite and config this run used.
OpenRouterHosted
curl https://openrouter.ai/api/v1/chat/completions \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"reflection/beam-1.0-501b-a23b","provider":{"order":["nebius/fp8"],"allow_fallbacks":false},"service_tier":"default","messages":[{"role":"user","content":"Hello"}]}'The request the benchmark sent, pinned to the provider that served it.
Configuration
Where the suite ran, and the token budget each task got.
- Hosted
- OpenRouter
- Nebius
- FP8
- 1M
- Not set
- v1.0.0
- 2026-10-16 16:23 UTC
{
"provider": {
"order": [
"nebius/fp8"
],
"allow_fallbacks": false
},
"service_tier": "default"
}| Task | Reasoning | Max output per call |
|---|---|---|
| Agents | 4.1Khigh | 12.3K |
| Structured Extraction | 4.1Khigh | 12.3K |
| Coding | 4.1Khigh | 12.3K |
| Grounded Answers | 4.1Khigh | 12.3K |
| Instruction Following | 4.1Khigh | 12.3K |
| Business Logic | 4.1Khigh | 12.3K |
| Security & permissions | 4.1Khigh | 12.3K |
| Diagnosis & investigation | 4.1Khigh | 12.3K |