NVIDIAAuto-detectedByteDance-Seed/Seed-OSS-36B-Instruct
Best configuration ranks #338 of 610 on the BenchLeader Index at 48.0 ±4.9. Last measured 20 Aug 2025stale: no new result in six months. Released 20 Aug 2025.
- Blended price
- $0.300/M
- $0.210 in · $0.570 out
- Output speed
- 26 tok/s
- First answer
- 80 s
- Context
- 262k
- Overall index48
- Reasoning53
- Agents & tools40
- Maths54
- Knowledge39
- Instruction following44
- Long context57
- Composite44
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| GPQA Diamond | 71.5%#153 | Epoch AI Benchmarking Hub |
| GPQA Diamond (AA)not in index | 72.6%#267 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 9.9%#260 | Artificial Analysis |
Agents & tools
| Benchmark | default | Source |
|---|---|---|
| Terminal-Bench Hard | 6.8%#240 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 49.4%#192 | Artificial Analysis |
Maths
| Benchmark | default | Source |
|---|---|---|
| OTIS Mock AIME | 67.5%#132 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| AA-Omniscience | -52.5#336 | Artificial Analysis |
Instruction following
| Benchmark | default | Source |
|---|---|---|
| IFBench | 41.9%#236 | Artificial Analysis |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 61.3%#210 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| AA Intelligence Index | 12.1#271 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0003 | – | 1.5 min |
| Summarise a 30-page report | 12,000 / 600 | $0.0029 | – | 1.7 min |
| Code edit | 6,000 / 1,500 | $0.0021 | – | 2.3 min |
| Agentic coding session | 60,000 / 4,000 | $0.015 | – | 3.9 min |
| Structured extraction | 2,000 / 200 | $0.0005 | – | 1.5 min |
See also
Data as of 9 Sept 2026. Compare with another model.