OpenAI
OpenAI: gpt-oss-20b
openai/gpt-oss-20bHugging Face ↗
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...
OpenAI: gpt-oss-20b usage over time
OpenRouter share of tokens, by day.
OpenAI: gpt-oss-20b holds 0.34% today, up from 0.24% at the start of the visible window.
Rank history
Adoption curve vs the pantheon
OpenAI: gpt-oss-20b's share plotted against days since first tracked, overlaid on other current leaders on the same axis. The question every new-model page asks: is this one going to make it?
Citable stat
As of 2026-07-14, OpenAI: gpt-oss-20b holds 0.3% of OpenRouter token share, up 0.09pp over 30 days. Source: Tidelines (OpenRouter).
Where to run it
Providers & pricing
Per-provider pricing, context, and uptime via OpenRouter. Sorted by output price.
| Provider | Input $/M | Output $/M | Context | Uptime |
|---|---|---|---|---|
| WandB | $0.03 | $0.13 | 131k | 99.82% |
| DekaLLM | $0.03 | $0.14 | 131k | 94.07% |
| DeepInfra | $0.03 | $0.14 | 131k | 100.00% |
| Novita | $0.04 | $0.15 | 131k | 99.97% |
| Phala | $0.04 | $0.15 | 131k | 99.90% |
| Amazon Bedrock | $0.07 | $0.15 | 131k | — |
| SiliconFlow | $0.04 | $0.18 | 131k | 98.92% |
| Parasail | $0.04 | $0.20 | 131k | 97.54% |
| Together | $0.05 | $0.20 | 131k | 59.29% |
| $0.07 | $0.25 | 131k | 63.08% | |
| Fireworks | $0.07 | $0.30 | 131k | 98.81% |
| Groq | $0.07 | $0.30 | 131k | 99.83% |