What this model supports
ReasoningYes
Tool callingYes
Structured outputYes
AttachmentsNo
Temperature controlYes
Open weightsYes
- Creator
- OpenAI
- Family
- gpt-oss
- Modalities
- Text
- License
- Not documented
- Release date
- 2025-08-05
- Model ID
openai/gpt-oss-20b
GPT OSS 20B
openai/gpt-oss-20bOpen GPT reasoning model for self-hosted agents and controllable deployments
Input modalities
Output modalities
Source-linkedgpt-ossOpen weightsReleased 2025-08-05
Context window131.072K
Maximum output32.768K
Provider offers26
Knowledge cutoffUndisclosed
Market availability
Report price →Provider pricing
| Provider | Provider model ID | Context | Input / 1M | Output / 1M | Cached / 1M | Features | All pricing |
|---|---|---|---|---|---|---|---|
 Nvidia | openai/gpt-oss-20b | 131.072K | $0 | $0 | — | R T S | View{
"input": 0,
"output": 0
} |
 Kenari | gpt-oss-20b | 131.072K | $0 | $0 | — | R T · | View{
"input": 0,
"output": 0
} |
 LMStudio | openai/gpt-oss-20b | 131.072K | $0 | $0 | — | R T · | View{
"input": 0,
"output": 0
} |
 OpenRouter | openai/gpt-oss-20b | 131.072K | $0.029 | $0.14 | — | R T S | View{
"input": 0.029,
"output": 0.14
} |
 IO.NET | openai/gpt-oss-20b | 64K | $0.03 | $0.14 | $0.015 | · T · | View{
"cache_read": 0.015,
"cache_write": 0.06,
"input": 0.03,
"output": 0.14
} |
 Weights & Biases | openai/gpt-oss-20b | 131.072K | $0.03 | $0.13 | $0.03 | R T S | View{
"cache_read": 0.03,
"input": 0.03,
"output": 0.13
} |
 Kilo Gateway | openai/gpt-oss-20b | 131.072K | $0.03 | $0.14 | — | R T · | View{
"input": 0.03,
"output": 0.14
} |
 Deep Infra | openai/gpt-oss-20b | 131.072K | $0.03 | $0.14 | — | R T S | View{
"input": 0.03,
"output": 0.14
} |
 SiliconFlow | openai/gpt-oss-20b | 131K | $0.04 | $0.18 | — | · T S | View{
"input": 0.04,
"output": 0.18
} |
 NovitaAI | openai/gpt-oss-20b | 131.072K | $0.04 | $0.15 | — | R · S | View{
"input": 0.04,
"output": 0.15
} |
 NanoGPT | openai/gpt-oss-20b | 128K | $0.04 | $0.15 | — | R · · | View{
"input": 0.04,
"output": 0.15
} |
 LLM Gateway | gpt-oss-20b | 131.072K | $0.04 | $0.15 | — | R T S | View{
"input": 0.04,
"output": 0.15
} |
 Clarifai | openai/chat-completion/models/gpt-oss-20b | 131.072K | $0.045 | $0.18 | — | R T · | View{
"input": 0.045,
"output": 0.18
} |
 FastRouter | openai/gpt-oss-20b | 131.072K | $0.05 | $0.2 | — | R T · | View{
"input": 0.05,
"output": 0.2
} |
 Together AI | openai/gpt-oss-20b | 131.072K | $0.05 | $0.2 | — | R T S | View{
"input": 0.05,
"output": 0.2
} |
 Databricks | databricks-gpt-oss-20b | 131.072K | $0.05 | $0.2 | — | R T S | View{
"input": 0.05,
"output": 0.2
} |
 Vercel AI Gateway | openai/gpt-oss-20b | 131.072K | $0.05 | $0.2 | — | R T S | View{
"input": 0.05,
"output": 0.2
} |
 Neon | gpt-oss-20b | 131.072K | $0.05 | $0.2 | — | R T S | View{
"input": 0.05,
"output": 0.2
} |
 Vertex | openai/gpt-oss-20b-maas | 131.072K | $0.07 | $0.25 | — | R T · | View{
"input": 0.07,
"output": 0.25
} |
 FrogBot | gpt-oss-20b | 131.072K | $0.07 | $0.2 | — | R T S | View{
"input": 0.07,
"output": 0.2
} |
 Pioneer | openai/gpt-oss-20b | 131.072K | $0.07 | $0.3 | $0.035 | R T S | View{
"cache_read": 0.035,
"cache_write": 0.07,
"input": 0.07,
"output": 0.3
} |
 Fireworks AI | accounts/fireworks/models/gpt-oss-20b | 131.072K | $0.07 | $0.3 | $0.035 | R T · | View{
"cache_read": 0.035,
"input": 0.07,
"output": 0.3
} |
 Groq | openai/gpt-oss-20b | 131.072K | $0.075 | $0.3 | $0.037 | R T S | View{
"cache_read": 0.0375,
"input": 0.075,
"output": 0.3
} |
 Hugging Face | openai/gpt-oss-20b | 131.072K | $0.1 | $0.5 | — | R T S | View{
"input": 0.1,
"output": 0.5
} |
 STACKIT | openai/gpt-oss-20b | 131.072K | $0.18 | $0.29 | — | R T · | View{
"input": 0.18,
"output": 0.29
} |
 Cloudflare Workers AI | @cf/openai/gpt-oss-20b | 128K | $0.2 | $0.3 | — | R T S | View{
"input": 0.2,
"output": 0.3
} |
Published benchmark results
Agentic Index
vs top 8 models
Coding Index
vs top 8 models
Design Arena: dataviz
vs top 8 models
Design Arena: website
vs top 8 models
Intelligence Index
vs top 8 models