# GPT OSS 20B

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs.

## Model Information

- **Organization**: [OpenAI](/llm.txt)
- **Slug**: gpt-oss-20b
- **Available at Providers**: 52
- **Release Date**: August 5, 2025

### Benchmark Scores
- HLE: 0.109
- GPQA: 0.715

## Providers

| Provider | Name | $ Input (per 1M) | $ Output (per 1M) | Free | Link |
|----------|------|-----------------|------------------|------|------|
| [AIHubMix](/llm/aihubmix.txt) | gpt-oss-20b | 0.11 | 0.55 |  | [View](https://aihubmix.com/model/gpt-oss-20b) |
| [AIHubMix](/llm/aihubmix.txt) | GPT-OSS-20B | 0.11 | 0.55 |  | [View](https://aihubmix.com/model/GPT-OSS-20B) |
| [AIMLAPI](/llm/aimlapi.txt) | GPT OSS 20B | 0.04 | 0.19 |  |  |
| [Chutes.ai](/llm/chutes.txt) | gpt-oss-20b | 0.04 | 0.15 |  |  |
| [FastRouter](/llm/fastrouter.txt) | OpenAI: GPT OSS 20B | 0.10 | 0.50 |  | [View](https://fastrouter.ai/models/openai/gpt-oss-20b) |
| [Fireworks AI](/llm/fireworks.txt) | OpenAI gpt-oss-20b | 0.07 | 0.30 |  |  |
| [Helicone](/llm/helicone.txt) | OpenAI GPT-OSS 20b | 0.05 | 0.20 |  | [View](https://www.helicone.ai/model/gpt-oss-20b) |
| [Groq](/llm/groq.txt) | GPT OSS 20B | 0.08 | 0.30 |  |  |
| [Nebius Token Factory](/llm/nebius.txt) | gpt-oss-20b | 0.05 | 0.20 |  | [View](https://huggingface.co/openai/gpt-oss-20b) |
| [Novita AI](/llm/novita.txt) | gpt-oss-20b | 0.04 | 0.15 |  |  |
| [SiliconFlow](/llm/siliconflow.txt) | gpt-oss-20b |  |  |  |  |
| [Cloudflare AI Gateway](/llm/cloudflareaigateway.txt) | GPT OSS 20B | 0.20 | 0.30 |  |  |
| [OVHcloud AI Endpoints](/llm/ovhcloud.txt) | gpt-oss-20b | 0.05 | 0.18 |  |  |
| [DeepInfra](/llm/deepinfra.txt) | gpt-oss-20b | 0.03 | 0.14 |  |  |
| [LMStudio](/llm/lmstudio.txt) | GPT OSS 20B | 0.00 | 0.00 | Yes |  |
| [IO.NET](/llm/ionet.txt) | GPT-OSS 20B | 0.03 | 0.14 |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | GPT OSS 20B |  |  |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | GPT-OSS 20B TEE |  |  |  |  |
| [OpenRouter](/llm/openrouter.txt) | gpt-oss-20b | 0.03 | 0.14 |  | [View](https://openrouter.ai/openai/gpt-oss-20b) |
| [Poe](/llm/poe.txt) | GPT-OSS-20B | 450.00 |  |  | [View](https://poe.com/gpt-oss-20b/api) |
| [Replicate](/llm/replicate.txt) | gpt-oss-20b |  |  |  |  |
| [Requesty](/llm/requesty.txt) |  | 0.10 | 0.50 |  |  |
| [Together AI](/llm/togetherai.txt) | OpenAI GPT-OSS 20B | 0.05 | 0.20 |  |  |
| [ValorGPT](/llm/valorgpt.txt) | gpt-oss-20b |  |  |  | [View](https://www.valorgpt.com/models/openai-gpt-oss-20b) |
| [ValorGPT](/llm/valorgpt.txt) | gpt-oss-20b |  |  | Yes | [View](https://www.valorgpt.com/models/openai-gpt-oss-20b-free) |
| [Vercel AI Gateway](/llm/vercel.txt) | gpt-oss-20b | 0.07 | 0.30 |  |  |
| [Yupp](/llm/yupp.txt) | gpt-oss-20b (OpenRouter) |  |  |  |  |
| [Routeway](/llm/routeway.txt) | OpenAI: GPT OSS 20B | 0.15 | 0.60 |  | [View](https://routeway.ai/models) |
| [LangDB](/llm/langdb.txt) | gpt-oss-20b |  |  |  | [View](https://langdb.ai/app/models) |
| [Kilo Code](/llm/kilocode.txt) | OpenAI: gpt-oss-20b | 0.03 | 0.14 |  | [View](https://kilo.ai/models/openai/gpt-oss-20b) |
| [Parasail](/llm/parasail.txt) | Gpt Oss 20b | 0.04 | 0.20 |  | [View](https://www.saas.parasail.io/pricing) |
| [Nvidia](/llm/nvidia.txt) | gpt-oss-20b |  |  |  | [View](https://build.nvidia.com/openai/gpt-oss-20b) |
| [RedPill](/llm/redpill.txt) | OpenAI: GPT OSS 20B | 0.04 | 0.15 |  |  |
| [302.AI](/llm/302ai.txt) | gpt-oss-20b | 0.10 | 0.50 |  | [View](https://302ai-en.apifox.cn/api-207705128) |
| [Weights & Biases](/llm/wandb.txt) | gpt-oss-20b |  |  |  |  |
| [Cloudflare Workers AI](/llm/cloudflareworkersai.txt) | GPT OSS 20B | 0.20 | 0.30 |  |  |
| [Google Vertex AI](/llm/googlevertex.txt) | Gpt Oss 20b |  |  |  |  |
| [Inference](/llm/inference.txt) |  |  |  |  |  |
| [Arena AI](/llm/arenaai.txt) |  |  |  |  |  |
| [Firmware](/llm/firmware.txt) | GPT OSS 20B | 0.07 | 0.20 |  |  |
| [Kilo Code](/llm/kilocode.txt) | OpenAI: gpt-oss-20b (free) | 0.00 | 0.00 | Yes |  |
| [Okara](/llm/okara.txt) | GPT-OSS 20B |  |  |  | [View](https://okara.ai/ai-models/gpt-oss-20b) |
| [Chats-LLM](/llm/chatsllm.txt) | OpenAI: gpt-oss-20b | 0.03 | 0.14 |  |  |
| [Blackbox AI](/llm/blackboxai.txt) | blackboxai/openai/gpt-oss-20b |  |  |  |  |
| [CometAPI](/llm/cometapi.txt) |  | 0.08 | 0.32 |  |  |
| [Qiniu](/llm/qiniuai.txt) | gpt-oss-20b |  |  |  |  |
| [ApiYI](/llm/apiyi.txt) | gpt-oss-20b |  |  |  |  |
| [WaveSpeed AI](/llm/wavespeed.txt) | gpt-oss-20b | 0.03 | 0.15 |  |  |
| [OpenRouter](/llm/openrouter.txt) | gpt-oss-20b (free) | 0.00 | 0.00 | Yes | [View](https://openrouter.ai/openai/gpt-oss-20b:free) |
| [Airforce API](/llm/airforce.txt) | gpt-oss-20b |  |  |  |  |
| [Writingmate](/llm/writingmate.txt) | OpenAI: gpt-oss-20b |  |  |  | [View](https://writingmate.ai/models/openai/gpt-oss-20b) |
| [LLM Stats](/llm/llmstats.txt) | GPT OSS 20B |  |  |  |  |

---

[← Back to all providers](/llm.txt)