unsloth / text-generation updated 6 months ago

Qwen3-Coder-Next-GGUF

Name: Qwen3-Coder-Next-GGUF
Author: unsloth

Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants.

Params

—

Context

—

Downloads 30d

196 K

Likes

841

Commercial use: allowed apache-2.0 Not gated GGUF View on Hugging Face ↗

Download history

daily snapshots · 56 days

▲ 14 K in the last 30 days (7.9%)

212 K174 K

Jul 28Aug 15Sep 3Sep 21

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
Qwen3-Coder-Next-IQ4_XS.gguf	IQ4_XS	42.7 GB	47.4 GB	❌ Won’t fit
Qwen3-Coder-Next-IQ4_NL.gguf	IQ4_NL	45.1 GB	50.1 GB	❌ Won’t fit
Qwen3-Coder-Next-BF16-00003-of-00004.gguf	GGUF	49.4 GB	54.9 GB	❌ Won’t fit
Qwen3-Coder-Next-Q4_1-00002-of-00003.gguf	Q4	49.7 GB	55.2 GB	❌ Won’t fit
Qwen3-Coder-Next-Q5_K_S-00002-of-00003.gguf	Q5_K_S	49.8 GB	55.2 GB	❌ Won’t fit
Qwen3-Coder-Next-Q8_0-00002-of-00003.gguf	Q8	49.8 GB	55.2 GB	❌ Won’t fit
Qwen3-Coder-Next-UD-Q5_K_M-00002-of-00003.gguf	Q5_K_M	49.9 GB	55.4 GB	❌ Won’t fit
Qwen3-Coder-Next-UD-Q6_K-00002-of-00003.gguf	Q6_K	50.0 GB	55.5 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

copy-paste, exact tags checked against the Hub

~ · ollama · Q4

$ ollama run qwen3-coder-next-gguf

# pin the quantization explicitly
$ ollama run qwen3-coder-next-gguf-q4

est. VRAM 55.2 GBon RTX 4090 · 24 GBJSON API →

Base model and the most-downloaded derivatives in the catalog.

Compare with any text-generation model