GPT-OSS 20B
by IO.NET · family gpt-oss · listed Aug 2025RTW
Open-weight GPT model for self-hosted reasoning and instruction-following workloads
Current rates USD per 1M tokens
Input$0.055
Output$0.194
Cache read$0.0275
Cache write—
Against the market
On the record
- Model id
openai/gpt-oss-20b- Context window
- 64K tokens
- Max output
- 4K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- Oct 2024
- Released
- 2025-08-05
- Last updated
- 2025-08-05
- Provider docs
- io.net/docs/guides/intelligence/io-intelligence
Back of the envelope
≈
Also listed at
- Nvidiafree in · free out
- QVACfree in · free out
- InferAllfree in · free out
- LMStudiofree in · free out
- Eden AI$0.009 in · $0.045 out
- SOKKAN Router$0.015 in · $0.075 out
- Kilo Gateway$0.018 in · $0.09 out
- OpenRouter$0.018 in · $0.09 out
- NanoGPT$0.018 in · $0.09 out
- Vercel AI Gateway$0.03 in · $0.14 out
- CoreWeave$0.03 in · $0.13 out
- Deep Infra$0.03 in · $0.14 out
- Kenari$0.036 in · $0.154 out
- DevPass (LLM Gateway)$0.04 in · $0.19 out
- Inference APIs$0.04 in · $0.19 out