AtomicChat / text-generation updated 1 month ago

Ling-3.0-flash-GGUF

Name: Ling-3.0-flash-GGUF
Author: AtomicChat

Quantizations of inclusionAI/Ling-3.0-flash: 124B total, 5.1B active, hybrid linear attention (35 KDA blocks interleaved 5:1 with 7 gated MLA blocks) over a 512-expert MoE.

Params

—

Context

—

Downloads 30d

348 K

Likes

Commercial use: allowed mit Not gated GGUF View on Hugging Face ↗

Download history

daily snapshots · 24 days

351 K183 K

Aug 28Sep 5Sep 13Sep 20

Can you run it?

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
Ling-3.0-flash-AD-IQ1_S.gguf	IQ1_S	32.4 GB	36.1 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ1_M.gguf	IQ1_M	36.5 GB	40.6 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ2_XXS.gguf	IQ2_XXS	39.2 GB	43.7 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ3_S-00001-of-00002.gguf	IQ3_S	44.6 GB	49.6 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ2_M-00001-of-00002.gguf	IQ2_M	44.7 GB	49.7 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ2_S-00001-of-00002.gguf	IQ2_S	44.7 GB	49.7 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ2_XS.gguf	IQ2_XS	44.7 GB	49.7 GB	❌ Won’t fit
Ling-3.0-flash-AD-IQ3_M-00001-of-00002.gguf	IQ3_M	45.0 GB	49.9 GB	❌ Won’t fit

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

Run it

copy-paste, exact tags checked against the Hub

~ · ollama · IQ1_M

$ ollama run ling-3-0-flash-gguf

# pin the quantization explicitly
$ ollama run ling-3-0-flash-gguf-iq1_m

est. VRAM 40.6 GBon RTX 4090 · 24 GBJSON API →

Specifications

Licence: mit
First seen on the Hub: 2026-08-04
Training datasets: undisclosed
Added to our catalog: 2026-08-28

Compare with any text-generation model