- Context
- 128K tokens
- Max output
- —
- Input / 1M
- —
- Output / 1M
- —
Model Family
Qwen3
Qwen3 is an open-weight Qwen language-model family with dense and mixture-of-experts variants spanning small local models through large flagship models.
Official documentationVerified Oct 4, 2026
Models
3
Provider
Alibaba Cloud
API access
—
Context range
32K–128K tokens
Input price range
—
Output price range
—
Overview
The family supports hybrid thinking modes, multilingual use, coding and agentic workflows. iilib initially represents it with 4B, 30B-A3B and 235B-A22B models rather than every published size and quantization.
Models in this family
Current indexed models and their documented specifications.
Qwen3-235B-A22B
Open weightsApr 29, 2025
0 mapped APIs
Qwen3-30B-A3B
Open weightsApr 29, 2025
- Context
- 128K tokens
- Max output
- —
- Input / 1M
- —
- Output / 1M
- —
0 mapped APIs
Qwen3-4B
Open weightsApr 29, 2025
- Context
- 32K tokens
- Max output
- —
- Input / 1M
- —
- Output / 1M
- —
0 mapped APIs
| Model | Released | Context | Max output | Input | Output | API |
|---|---|---|---|---|---|---|
| Qwen3-235B-A22B Open weights | Apr 29, 2025 | 128K tokens | — | — | — | — |
| Qwen3-30B-A3B Open weights | Apr 29, 2025 | 128K tokens | — | — | — | — |
| Qwen3-4B Open weights | Apr 29, 2025 | 32K tokens | — | — | — | — |