Currently listed through these providers:
Model details
Qwen3-Coder 30B-A3B Instruct
Qwen3-Coder 30B-A3B Instruct is the 30B member of the Qwen3-Coder family, offered in an instruction-oriented form for software-development workflows. Its A3B designation is part of the documented model name, but the supplied evidence does not establish an architecture, expert layout, training method, training scale, or benchmark performance.
The strongest verified practical signal is efficiency: the family presentation describes this variant as streamlined while maintaining performance, although no comparative efficiency metrics are supplied. A community-quantized Ollama build also shows that it can run locally and be connected to coding agents such as Claude Code, OpenCode, Hermes Agent, and OpenClaw. This makes it a plausible option for users prioritizing local deployment and agent-assisted programming over unsupported claims about frontier results or specialized training.
Quick Info
Powered by- Provider
- Pendra
- Model key
- qwen3-coder:30b
- Release date
- Apr 1, 2025
- Last updated
- Apr 1, 2025
- Knowledge cutoff
- 2025-04
- Input modalities
- Output modalities
- Capabilities
Cost
A provider subscription or plan supersedes token-based pricing for this model.
Limits
- Output tokens
- 65,536 tokens
- Context window
- 262,144 tokens