gpt-oss-20b
Intelligence
9.04
#372 of 520 models
Output speed
168.3 tok/s
Measured generation rate
Cost / index task
$0.011
Measured evaluation workload
End-to-end latency
16ms
First answer 13ms
01 / Position
A model in the landscape
Compare this model with peers across intelligence, speed, cost, and individual evaluations.
02 / Evidence
Benchmark results
28 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
- $0.060
- Output / 1M
- $0.190
- Cache read / 1M
- —
- Cache write / 1M
- —
- Cache-hit discount
- —
Response & usage
- Output speed
- 168.3 tok/s
- First chunk
- 1ms
- First answer
- 13ms
- Output / task
- 18.1k
- Reasoning / task
- 16.1k
Evaluation workload
- Canonical input
- 269.18M
- Canonical output
- 69.79M
- Answer / task
- 2k
- Index time / task
- 1.5m
- Run cost
- $29.412
By prompt type
| Prompt | Output speed | First chunk | First answer | End to end |
|---|---|---|---|---|
| Medium | 196.7 tok/s | 0ms | 11ms | 13ms |
| Long | 168.3 tok/s | 1ms | 13ms | 16ms |
| Hundred K | 102.3 tok/s | 5ms | 24ms | 29ms |
| Medium Parallel | 169.3 tok/s | 0ms | 12ms | 15ms |
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-20b
- Tokenizer
- GPT
- Instruct type
- —
- Captured endpoints
- 13
| Provider | Quantization | Context | tok/s |
|---|---|---|---|
| Darkbloom | fp8 | 131.1k | — |
| AkashML | fp4 | 131.1k | — |
| DekaLLM | bf16 | 131.1k | — |
| CoreWeave | fp4 | 131.1k | — |
| DeepInfra | bf16 | 131.1k | — |
| Parasail | fp4 | 131.1k | — |
| Novita | fp4 | 131.1k | — |
| SiliconFlow | fp8 | 131.1k | — |
| Together | unknown | 131.1k | — |
| Amazon Bedrock | unknown | 131.1k | — |
| Amazon Bedrock | unknown | 131.1k | — |
| unknown | 131.1k | — | |
| Groq | unknown | 131.1k | — |
05 / Audit
Underlying fields
Inspect the captured model and routing records. Artificial Analysis destination links are omitted.