arcflare_
Code80B-A3B official

qwen3-coder-next

by Alibaba Qwen

Qwen3-Coder-Next: an 80B MoE with only 3B active, a 256K context, and training aimed at agent harnesses like Claude Code and Cline.

#code#agentic#moe#long-context
arcflare pull unsloth/Qwen3-Coder-Next-GGUF:Q4_K_M

[ 01 ] // quickstart

┌─ arcflare — zsh // tty1
$ 

[ 02 ] // readme

qwen3-coder-next

Qwen3-Coder-Next: an 80B MoE with only 3B active, a 256K context, and training aimed at agent harnesses like Claude Code and Cline.

Get it

arcflare pull unsloth/Qwen3-Coder-Next-GGUF:Q4_K_M
arcflare run qwen3-coder-next

ArcFlare reads the GGUF header and loads it at the largest context your free VRAM can actually hold. Use arcflare fit to see that before downloading anything else, or arcflare use opencode to code with it.

Highlights

  • 3B active per token: fast once loaded, but the weights need ~50 GB of memory
  • A good fit for 64 GB unified-memory machines
  • 256K native context

Sizes

  • 80b-a3b · 48.5GB

Other sizes live in sibling repos: swap the size in the repo name when pulling.

[ 03 ] // relatedmore →

glm4.7-flash

Z.ai

Code

GLM-4.7-Flash: a 30B-A3B MoE built for agentic coding. Only ~3B parameters active per token, so it runs at small-model speed.

size
30B-A3B
vram
~20 GB
license
MIT

83% of a 24 GB card

#code #agentic #moe

qwen3-coder

Alibaba Qwen

Code

Qwen3-Coder 30B-A3B: an agentic coding MoE with 256K context that runs at small-model speed on one GPU.

size
30B-A3B
vram
~19 GB
license
Apache-2.0

79% of a 24 GB card

#code #agentic #moe

deepseek-coder-v2-lite

DeepSeek

Code

DeepSeek-Coder-V2-Lite: a 16B mixture-of-experts coder with a long context. Cheap to run for its size.

size
16B
vram
~12 GB
license
DeepSeek License

50% of a 24 GB card

#code #moe #long-context