← All models

lean-coder-welter

Welterweight class · v1.0 · Qwen 3-Coder-Next base

Code generation - debugging, refactoring, code review

Verified vs llama.cppFree

80B total with 512 experts, only 3B active per token. A 48.7 GB model (Q4_K_M) that runs on 12 GB VRAM. Tuned for code generation, debugging, and software engineering.

Specifications

Total params

80B

Active per token

3B

Base model

Qwen3-Coder-Next

Architecture

MoE (512 experts)

Experts

512 experts, 10 active

Min VRAM / RAM

12 GB / 32 GB

Measured performance

Real numbers from our reference rig, not projections.

HardwarePrefillDecode
1x RTX 3090 (24 GB)12-23 tok/s14-20 tok/s
2x RTX 30908-17 tok/s10-18 tok/s

Download

Download 48.7 GB - single .lmpack, Q4_K_M
$ lean pull lean-coder-welter

Quick start (coming soon)

$ curl -sSf https://leanmodels.ai/install.sh | sh
$ lean pull lean-coder-welter
$ lean run lean-coder-welter

Single binary, 15 MB. No Python, no Docker, no cloud dependency.

The installer is not live yet — binaries ship with our first release.