Qwen3-Coder-Next

Alibaba llm

Open-weight model built for coding agents: 80B total, 3B active per token, 256K context.

overall ⌄

36.3

68 of 93 ranked

price · $/M tokens

self-hosted

open weights, no first-party API

to run it yourself ⌄

48 GB

minimum · 192 GB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmark	published	measured
SWE-bench	70.6	—
SWE-bench Pro	44.3	—

No recorded lineage — root or standalone model.