Model prices / OpenAI / GPT OSS 20B
GPT OSS 20BOpen weights
OpenAI · released 2025-08-05
No price from the maker itself is on record for GPT OSS 20B. The lowest of 41 providers is Eden AI: $0.009 per 1M input tokens and $0.045 per 1M output tokens (as of 2026-11-12). Context window: 131K tokens, up to 32.8K output tokens per request.
Price by providerUSD per 1M tokens
| Provider | Input | Output | Cached | Notes | |
|---|---|---|---|---|---|
EEden AILowestgreenference/ | $0.009 | $0.045 | $0.0045 | Lowest | Official docs |
KKilo Gatewayopenai/ | $0.018 | $0.09 | $0.009 | Official docs | |
NNanoGPTopenai/ | $0.018 | $0.09 | $0.009 | Official docs | |
EEden AIflexai/ | $0.02 | $0.1 | $0.003 | Official docs | |
 OpenRouteropenai/ | $0.03 | $0.13 | $0.03 | Official docs | |
 CoreWeaveopenai/ | $0.03 | $0.13 | $0.03 | Official docs | |
 Deep Infraopenai/ | $0.03 | $0.14 | — | Official docs | |
EEden AIdeepinfra/ | $0.03 | $0.14 | — | Official docs | |
 Vercel AI Gatewayopenai/ | $0.03 | $0.14 | — | Official docs | |
MMerge Gatewayopenai/ | $0.04 | $0.15 | $0.02 | Official docs | |
KKenarigpt-oss-20b | $0.036 | $0.154286 | $0.003714 | Official docs | |
CCortecsgpt-oss-20b | $0.045 | $0.167 | — | Official docs |
40 more listings
| Provider | Input | Output | Cached | Notes | |
|---|---|---|---|---|---|
 SiliconFlowopenai/ | $0.04 | $0.18 | — | Official docs | |
CClarifaiopenai/ | $0.045 | $0.18 | — | Official docs | |
EEden AIovhcloud/ | $0.05 | $0.18 | — | Official docs | |
OOVHcloud AI Endpointsgpt-oss-20b | $0.05 | $0.18 | — | Official docs | |
 DevPass (LLM Gateway)gpt-oss-20b | $0.04 | $0.19 | $0.01 | Official docs | |
LLLM Gatewayconsensusprotocol/ | $0.04 | $0.19 | $0.01 | Official docs | |
 IO.NETopenai/ | $0.055 | $0.194 | $0.0275 | Official docs | |
HHeliconegpt-oss-20b | $0.05 | $0.2 | — | Official docs | |
FFrogBotgpt-oss-20b | $0.07 | $0.2 | — | Official docs | |
FFastRouteropenai/ | $0.07 | $0.25 | $0.007 | Official docs | |
 Amazon Bedrockopenai. | $0.07 | $0.3 | — | Official docs | |
| Amazon Bedrockopenai. | $0.07 | $0.3 | — | Official docs | |
DDatabricksdatabricks-gpt-oss-20b | $0.07 | $0.3 | — | Official docs | |
NNeongpt-oss-20b | $0.07 | $0.3 | — | Official docs | |
OOCI Generative AIopenai. | $0.07 | $0.3 | — | Official docs | |
 Ollama Cloudgpt-oss: | $0.07 | $0.3 | $0.035 | Official docs | |
PPrime Inferenceopenai/ | $0.07 | $0.3 | — | Official docs | |
EEden AIdatabricks/ | $0.07 | $0.30002 | $0.007 | Official docs | |
EEden AIdatabricks/ | $0.07 | $0.30002 | $0.007 | Official docs | |
EEden AIgroq/ | $0.075 | $0.3 | $0.0375 | Official docs | |
 Groqopenai/ | $0.075 | $0.3 | $0.0375 | Official docs | |
TTempr Gatewaygroq/ | $0.075 | $0.3 | $0.0375 | Official docs | |
 AI/ML APIopenai/ | $0.07935 | $0.3174 | $0.039675 | Official docs | |
| Amazon Bedrockus-gov. | $0.084 | $0.36 | — | Official docs | |
SSTACKITopenai/ | $0.18 | $0.29 | — | Official docs | |
 Cloudflare Workers AI@cf/ | $0.2 | $0.3 | — | Official docs | |
 DigitalOceanopenai-gpt-oss-20b | $0.05 | $0.45 | $0.01 | Official docs | |
EEden AIcloudflare/ | $0.2 | $0.3 | — | Official docs | |
RRegolo AIgpt-oss-20b | $0.11215 | $0.47103 | — | Official docs | |
LLowRouterauto/ | $0.113 | $0.4745 | — | Official docs | |
 Hugging Faceopenai/ | $0.1 | $0.5 | — | Official docs | |
LLLM Gatewaygroq/ | $0.1 | $0.5 | — | Official docs | |
OOppergpt-oss-20b | $0.11622 | $0.488124 | — | Official docs | |
TThinking Machinesopenai/ | $0.18 | $0.45 | $0.036 | Official docs | |
IImpossiblDeprecatedfireworks/ | $0.07 | $0.3 | $0.035 | Deprecated | Official docs |
IImpossiblDeprecatedgroq/ | $0.075 | $0.3 | $0.0375 | Deprecated | Official docs |
 LMStudioopenai/ | $0 | $0 | — | Official docs | |
 Nvidiaopenai/ | $0 | $0 | — | Official docs | |
QQVACgpt-oss-20b | $0 | $0 | — | Official docs | |
QQiniuDeprecatedgpt-oss-20b | — | — | — | Deprecated | Official docs |
Providers are those listed by models.dev. This site has not vetted them one by one; each provider's own page has the binding price.
About
Open GPT reasoning model for self-hosted agents and controllable deployments
Description from models.dev.
Specs
- Type
- Text
- Context
- 131K
- Max output
- 32.8K
- Input
- Text
- Output
- Text
- Reasoning
- Yes
- Tool calling
- Yes
- Structured output
- Yes
- Knowledge cutoff
- 2024-06
- Open weights
- Yes
- Released
- 2025-08-05
Source
Prices and specs come from models.dev (MIT licence), updated 2026-11-12. Each provider's name links to its official docs. A provider priced at 0 usually includes the model in a subscription; it is not necessarily a free API.