gpt-oss-20b
$0.018 per 1M input tokens and $0.09 per 1M output tokens · 131K context · tools, reasoning, JSON, open weights.
Prices checked Oct 21, 2026 · tracked by LLM Cost Index since Oct 21, 2026
What gpt-oss-20b costs per month
Estimated monthly bill for typical workloads, using the cached-input price for repeated context. Try your own numbers in the cost calculator.
| Workload | Per month |
|---|---|
| Customer chatbot 1,000 conversations a day, 8 turns each | $12.10 |
| Document Q&A (RAG) 500 questions a day over retrieved documents | $2.21 |
| Coding / tool-using agent 200 tasks a day, long tool-heavy contexts | $6.21 |
| Content generation 30 long articles a day | $0.251 |
| Classification / extraction 100,000 short items a day | $20.25 |
Price history
We started tracking gpt-oss-20b on Oct 21, 2026; the chart appears as soon as there is a second data point. See all recent price changes.
Where to run gpt-oss-20b: 11 providers
Darkbloom is the cheapest at $0.018 input / $0.09 output per 1M tokens. Uptime is the average over the last 30 days as reported by OpenRouter; see provider reliability.
| Provider | Input / 1M | Output / 1M | Context | Uptime | Notes |
|---|---|---|---|---|---|
| Darkbloom | $0.018 | $0.09 | 131K | 99.9% | fp8 |
| AkashML | $0.02 | $0.10 | 131K | 99.9% | fp4 |
| DekaLLM | $0.029 | $0.14 | 131K | 100% | bf16 |
| CoreWeave | $0.03 | $0.13 | 131K | 100% | fp4 |
| DeepInfra | $0.03 | $0.14 | 131K | 99.2% | bf16 |
| Parasail | $0.03 | $0.15 | 131K | 98.3% | fp4 |
| SiliconFlow | $0.04 | $0.18 | 131K | 90.9% | fp8 |
| Amazon Bedrock | $0.07 | $0.15 | 131K | 99.9% | — |
| Amazon Bedrock | $0.07 | $0.15 | 131K | 100% | — |
| $0.07 | $0.25 | 131K | 97.1% | — | |
| Groq | $0.075 | $0.30 | 131K | 100% | — |
How to pay less for gpt-oss-20b
For a customer chatbot (1,000 conversations a day, 8 turns each) that costs $14.04 a month at list price without caching:
| Option | Per month | You save |
|---|---|---|
| Use prompt caching (60% of input cached) | $12.10 | $1.94 (14%) |
Other workloads: cut-your-bill tool for gpt-oss-20b.
Specifications
| Model ID | openai/gpt-oss-20b |
| Input | text |
| Output | text |
| Tool calling | Yes |
| Structured output | Yes |
| Reasoning mode | Yes |
| Open weights | Yes — openai/gpt-oss-20b |
| Knowledge cutoff | 2024-06-30 |
| Released | Aug 5, 2025 |
FAQ
- How much does gpt-oss-20b cost?
- $0.018 per million input tokens and $0.09 per million output tokens; cached input costs $0.0090 per million. Prices as of Oct 21, 2026.
- What is the context window of gpt-oss-20b?
- 131K tokens, with up to 33K output tokens per response.
- Does gpt-oss-20b support images, tools and structured output?
- Images as input: no. Tool calling: yes. Structured (JSON) output: yes. Reasoning mode: yes.
More from OpenAI: all OpenAI models · how we collect prices