GPT OSS 20B
by OpenAIChat
Open-weight GPT model for self-hosted reasoning and instruction-following workloads
- Context
- 131K tokens
- Max output
- 65,536 tokens
- Parameters
- 20.9B total
- Released
- 2025-08-05
- Weights
- Open
- Licence
- apache-2.0
openai/gpt-oss-20b · also seen as databricks-gpt-oss-20b, fireworks/gpt-oss-20b, groq/gpt-oss-20b, @cf/openai/gpt-oss-20b, +19 more
Specifications
Full facts as recorded, incl. what is unknown.
- Released
- 2025-08-05
- Weights
- Opennot gated
- License
- apache-2.0
- Input
- Text
- Output
- Text
- Context window
- 131,072 max across endpoints
- Max output
- 65,536
- Providers
- 42 + 20 gateways, not counted as providers
- Epoch Capabilities Index
- 137.8 ± 3.3 against price
- Direct price range
- $0.0145–$0.20 in · $0.07–$0.50 out per 1M tokens · 37 of 42 providers have a comparable price
More details
- Last updated
- 2026-04-16
- Knowledge cutoff
- 2024-10
- Weights link
- Hugging Face · openai/gpt-oss-20b
- Parameters
- 20.9B total
- Developer country
- United States of America
- Training compute
- 5.5×10²³ FLOP
Capabilities
Across all endpoints: supported if any endpoint supports it.
- Tool calling
- Parallel tools
- Structured output
- Reasoning
- Image input
- PDF input
- Audio input
- Audio outputunknown
- Video input
- Prompt caching
- Web search
- Computer useunknown
- System messages
- Assistant prefillunknown
- Temperature
- Streaming
- Attachments
Endpoints
One row per provider and variant; gateways are listed separately.
Loading endpoints…
Price history
Atom feedNo change since 19 Oct 2026. The cheapest direct price has been $0.0145 in / $0.07 out per 1M tokens at Darkbloom on every data refresh since then.
Tracked from 19 Oct 2026, the earliest refresh this model's headline price can be rebuilt from. Headline-eligible direct prices only, so a price the sanity check flags never draws a spike. Series as JSON
Cost calculator
Loading endpoints…
Similarly priced models
Blended rate per 1M tokens, each at its own cheapest provider (10M input : 2M output mix).
| Model | Blended/1M | In/1M | Out/1M | Context | vs GPT OSS 20B |
|---|---|---|---|---|---|
| Mistral NemoMistral AI | $0.0292 | $0.029 | $0.03 | 131K | −3% cheaper |
| Ling 3.0 FlashinclusionAI | $0.028 | $0.021 | $0.063 | 262K | −7% cheaper |
| Granite 4.0 H MicroIBM | $0.0328 | $0.017 | $0.112 | 131K | +9% |
| Llama 3.2 3B InstructMeta | $0.0333 | $0.03 | $0.05 | 131K | +11% |
| Sarvam-30BSarvam AI | $0.0333 | $0.02 | $0.10 | 65.5K | +11% |
| Nex-N2.5-MiniNex AGI | $0.0375 | $0.025 | $0.10 | 262K | +25% |
| Mistral Nemo Instruct 2407Mistral AI | $0.0217 | $0.02 | $0.03 | 131K | −28% cheaper |
| Ministral 3BMistral AI | $0.04 | $0.04 | $0.04 | 131K | +33% |
More from OpenAI
Newest OpenAI models in the catalogue. All OpenAI models
| Model | Released | Context | Listed price |
|---|---|---|---|
| GPT-6.1 Sol Ultrafast | 2026-10-08 | 1.05M | from $12 |
| GPT-6 Luna Decisions | 2026-10-06 | 1.05M | unpriced |
| GPT-6.1 Sol | 2026-09-29 | 1.05M | from $2.00 |
| GPT-6.1 Sol Pro | 2026-09-29 | 1.05M | from $2.00 |
| GPT-6 Luna | 2026-09-22 | 1.05M | from $0.10 |
| GPT-6 Sol | 2026-09-22 | 1.05M | from $2.00 |
| GPT-6 Sol Pro | 2026-09-22 | 1.05M | from $2.00 |
| GPT-6 Luna Pro | 2026-09-22 | 1.05M | from $0.10 |