gpt-oss-120b
Intelligence
12.35
#312 of 520 models
Output speed
181.7 tok/s
Measured generation rate
Cost / index task
$0.107
Measured evaluation workload
End-to-end latency
15ms
First answer 12ms
01 / Position
A model in the landscape
Compare this model with peers across intelligence, speed, cost, and individual evaluations.
02 / Evidence
Benchmark results
31 distinct evaluations grouped by topic. Explore cross-model charts or switch to the full results list.
Domain capability
03 / Operation
Speed, tokens & price
Prices are per million tokens; latency and task costs describe the captured evaluation rather than a universal provider guarantee.
Token pricing
- Input / 1M
- $0.150
- Output / 1M
- $0.595
- Cache read / 1M
- $0.125
- Cache write / 1M
- —
- Cache-hit discount
- 47.5%
Response & usage
- Output speed
- 181.7 tok/s
- First chunk
- 1ms
- First answer
- 12ms
- Output / task
- 27k
- Reasoning / task
- 15.1k
Evaluation workload
- Canonical input
- 481.41M
- Canonical output
- 84.17M
- Answer / task
- 11.9k
- Index time / task
- 2.4m
- Run cost
- $122.292
By prompt type
| Prompt | Output speed | First chunk | First answer | End to end |
|---|---|---|---|---|
| Medium | 187.1 tok/s | 1ms | 11ms | 14ms |
| Long | 181.7 tok/s | 1ms | 12ms | 15ms |
| Hundred K | 155.9 tok/s | 4ms | 17ms | 20ms |
| Medium Parallel | 189.8 tok/s | 1ms | 11ms | 14ms |
04 / Access
License & availability
Recorded license
Apache 2.0
The publisher-linked license text defines permitted use and redistribution. Review it before using the model.
- Weight access
- Open weights
- Classification
- Permissive
- Reasoning
- Yes
Routing and providers
- OpenRouter ID
- openai/gpt-oss-120b
- Tokenizer
- GPT
- Instruct type
- —
- Captured endpoints
- 24
| Provider | Quantization | Context | tok/s |
|---|---|---|---|
| AkashML | bf16 | 131.1k | — |
| CoreWeave | fp4 | 131.1k | — |
| DekaLLM | bf16 | 131.1k | — |
| DeepInfra | bf16 | 131.1k | — |
| Crusoe | bf16 | 131.1k | — |
| Novita | fp4 | 131.1k | — |
| Mancer 2 | fp8 | 131.1k | — |
| DigitalOcean | unknown | 128k | — |
| unknown | 131.1k | — | |
| BaseTen | fp4 | 128.1k | — |
| BaseTen | fp4 | 128.1k | — |
| Parasail | fp4 | 131.1k | — |
| SambaNova | unknown | 131.1k | — |
| Amazon Bedrock | unknown | 131.1k | — |
| Nebius | fp4 | 131.1k | — |
| Amazon Bedrock | unknown | 131.1k | — |
| DeepInfra | bf16 | 131.1k | — |
| SiliconFlow | fp8 | 131.1k | — |
| Phala | unknown | 131.1k | — |
| Together | unknown | 131.1k | — |
| Groq | unknown | 131.1k | — |
| Mara | unknown | 131.1k | — |
| DeepInfra | fp8 | 131.1k | — |
| Cerebras | fp16 | 131.1k | — |
05 / Audit
Underlying fields
Inspect the captured model and routing records. Artificial Analysis destination links are omitted.