gpt-oss-120b pricing and routing profile
gpt-oss-120b from OpenAI lists at $0.15 per million input tokens and $0.60 per million output tokens, with a 131,072-token context window. Prices from OpenRouter's catalogue on 2026-09-28.
Tool callingStructured outputReasoning controlLog probabilities
Cost per 1,000 requests
| Request shape | List cost |
|---|---|
| Classification or routing (600 tokens in, 10 out) | $0.096 |
| Chat reply (2,000 in, 400 out) | $0.54 |
| Retrieval-augmented answer (8,000 in, 500 out) | $1.50 |
Before prompt caching. Try your own shape in the cost calculator.
Can Gimbix route to it first?
Yes. 9 of 24 tracked hosts return token log probabilities for gpt-oss-120b, which is the confidence signal Gimbix uses to let a cheaper model answer first.
Measured by Gimbix
On 400 real BANKING77 customer messages, gpt-oss-120b answered 329 correctly (95% interval 78.2%–85.7%), with 2 invalid answers, at $0.0229 per 1,000 requests and a median of 1,280 ms. See all 22 models.
Who serves gpt-oss-120b
| Host | Input / output per 1M | Log probabilities | Uptime, last 30 min |
|---|---|---|---|
| AkashML | $0.03 / $0.17 | yes | 100.0% |
| CoreWeave | $0.03 / $0.17 | yes | 100.0% |
| DekaLLM | $0.03 / $0.18 | yes | 99.0% |
| DeepInfra | $0.037 / $0.17 | no | 99.9% |
| Crusoe | $0.05 / $0.25 | no | 100.0% |
| Novita | $0.05 / $0.25 | yes | 100.0% |
| Mancer 2 | $0.05 / $0.30 | yes | 100.0% |
| DigitalOcean | $0.06 / $0.42 | yes | 100.0% |
| $0.09 / $0.36 | no | 63.2% | |
| BaseTen | $0.10 / $0.50 | no | 99.9% |
| BaseTen | $0.10 / $0.50 | no | 99.9% |
| Parasail | $0.10 / $0.75 | yes | 100.0% |
| SambaNova | $0.14 / $0.95 | no | 94.8% |
| Amazon Bedrock | $0.15 / $0.60 | no | — |
| Nebius | $0.15 / $0.60 | no | 99.7% |
| Amazon Bedrock | $0.15 / $0.60 | no | 100.0% |
| DeepInfra | $0.15 / $0.60 | no | 99.8% |
| SiliconFlow | $0.15 / $0.60 | no | 100.0% |
| Phala | $0.15 / $0.60 | yes | 100.0% |
| Together | $0.15 / $0.60 | no | 92.5% |
| Groq | $0.15 / $0.60 | no | 99.1% |
| Mara | $0.15 / $0.75 | yes | 77.1% |
| DeepInfra | $0.20 / $0.95 | no | 95.3% |
| Cerebras | $0.35 / $0.75 | yes | 100.0% |
A host that lists log probabilities may still omit them; Gimbix checks each one before relying on it.
Cheaper models that can answer first
Models priced below gpt-oss-120b that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.
| Model | Input / output per 1M | Cheaper on a chat reply |
|---|---|---|
| Qwen3 VL 8B Instruct | $0.117 / $0.455 | 23% |
| DeepSeek V4 Flash 0423 | $0.14 / $0.28 | 27% |
| Qwen3 VL 32B Instruct | $0.104 / $0.416 | 31% |
| Ternary Bonsai 2 27B | $0.075 / $0.50 | 35% |
| Nemotron 3 Super | $0.08 / $0.45 | 37% |
| Gemma 3 27B | $0.08 / $0.45 | 37% |
Compare gpt-oss-120b
- gpt-oss-120b vs Claude Opus 5.5
- gpt-oss-120b vs Claude Sonnet 5
- gpt-oss-120b vs Claude Fable 5.1
- gpt-oss-120b vs Claude Opus 5
- gpt-oss-120b vs GPT-6 Sol
- gpt-oss-120b vs GPT-6 Astra
- gpt-oss-120b vs GPT-5.6 Sol
- gpt-oss-120b vs GPT-5.6 Terra
- gpt-oss-120b vs GPT-5.5
- gpt-oss-120b vs GPT-5.4
- gpt-oss-120b vs Gemini 3.1 Pro Preview
- gpt-oss-120b vs Gemini 3.8 Flash
- gpt-oss-120b vs Grok 4.7
- gpt-oss-120b vs Qwen3.8 Max (0902)
- gpt-oss-120b vs Kimi K3
- gpt-oss-120b vs GLM 5.3
- gpt-oss-120b vs DeepSeek V4 Pro 0423
- gpt-oss-120b vs Mistral Large 3 2512
- gpt-oss-120b vs Nova Premier 1.0
- gpt-oss-120b vs Command A
- gpt-oss-120b vs GPT-6 Luna
- gpt-oss-120b vs GPT-5.6 Luna
- gpt-oss-120b vs GPT-5.4 Mini
- gpt-oss-120b vs GPT-5.4 Nano
- gpt-oss-120b vs GPT-4.1 Mini
- gpt-oss-120b vs gpt-oss-20b
- gpt-oss-120b vs Claude Haiku 4.5
- gpt-oss-120b vs Gemini 3.5 Flash Lite
- gpt-oss-120b vs Gemma 4 31B
- gpt-oss-120b vs DeepSeek V4.1 Flash
- gpt-oss-120b vs DeepSeek V4 Flash 0731
- gpt-oss-120b vs GLM 5.3 Flash
- gpt-oss-120b vs Hy4 preview
- gpt-oss-120b vs MiMo-V2.6-Flash
- gpt-oss-120b vs Nemotron 3.5 Lightning
- gpt-oss-120b vs Nemotron 3 Ultra
- gpt-oss-120b vs Qwen3.8 Flash
- gpt-oss-120b vs Qwen3.7 Flash
- gpt-oss-120b vs Qwen3.8 27B
- gpt-oss-120b vs MiniMax M3
- gpt-oss-120b vs Kimi K2.6
- gpt-oss-120b vs Mistral Small 4
- gpt-oss-120b vs Ministral 3 8B 2512
- gpt-oss-120b vs Llama 4 Maverick
- gpt-oss-120b vs Llama 3.3 70B Instruct
- gpt-oss-120b vs Nova Lite 1.0
- gpt-oss-120b vs Nova Micro 1.0
- gpt-oss-120b vs Command R7B (12-2024)
- gpt-oss-120b vs Phi 4
Questions
How much does gpt-oss-120b cost?
gpt-oss-120b lists at $0.15 per million input tokens and $0.60 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $0.54 per 1,000 requests.
What is the context window of gpt-oss-120b?
131,072 tokens.
Can a cheaper model replace gpt-oss-120b?
Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match gpt-oss-120b on one half of your requests and make at most three answers worse on the other.