Alibaba
Qwen3 8B
Qwen instruction model for multilingual chat, reasoning, and tool use
Model facts
- Context window
- 131K tokens
- Maximum output
- 8K tokens
- Cheapest paid input
- $0.035 per 1M tokens
- Cheapest paid output
- $0.138 per 1M tokens
- Open weights
- yes
- Providers
- 6
Capabilities and modalities
reasoning, tool calling, temperature control, open weights, text.
Local hardware estimate
At 4K context: Q4_K_M 6.0 GiB, Q8_0 9.8 GiB, F16 17.0 GiB working memory.
- Parameters
- 8.2 billion
- Architecture
- qwen3
- Model license
- apache-2.0
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| Reasoning | 8.3 | 9.0 | 19.0 | not measured |
| Non-reasoning | 4.8 | not measured | 24.3 | not measured |
Observed price history
- 2026-08-24: $0.035 input / $0.138 output per 1M tokens
API providers
- NovitaAI — model id qwen/qwen3-8b-fp8; $0.035 input / $0.138 output per 1M tokens; provider documentation
- Alibaba (China) — model id qwen3-8b; $0.072 input / $0.287 output per 1M tokens; provider documentation
- Pioneer — model id Qwen/Qwen3-8B; $0.2 input / $0.2 output per 1M tokens; provider documentation
- OpenRouter — model id qwen/qwen3-8b; $0.117 input / $0.455 output per 1M tokens; provider documentation
- Alibaba — model id qwen3-8b; $0.18 input / $0.7 output per 1M tokens; provider documentation
- NanoGPT — model id qwen/Qwen3-8B; $0.47 input / $0.47 output per 1M tokens; provider documentation