Model Terminal

Qwen3-Coder-480B-A35B-Instruct

Qwen3-Coder-480B-A35B-Instruct is Alibaba's Qwen team's flagship open-weight coding model — a 480B-parameter mixture-of-experts LLM with 35B active parameters designed for code generation, repository-scale understanding, tool use, and agentic coding workflows. It supports a native 256K context window (extendable to 1M via YaRN) and targets performance comparable to frontier closed models on coding benchmarks. Source: supabase.

Value score
Not scored
Context
tokens
Max output
tokens
Price

Capability radar

Identity

Developer
Qwen
Openness
Modalities
Release
Knowledge cutoff
Deprecation
API docs

Benchmarks

No benchmark scores yet.

Compare nearby

Pricing

Input / 1M
Output / 1M
Speed