unsloth / image-text-to-text updated 2 months ago

gemma-4-12b-it-GGUF

Jun 9 Update: Added MTP support. See our MTP Guide. Gemma 4 12B can now be run and fine-tuned in Unsloth Studio. Read our guide. See all versions of Gemma 4 (GGUF, 16-bit etc.) in our collection. Example of Gemma 4 E4B (4-bit GGUF) running in Unsloth Studio with tool-calling:

Params
12.0 B
Context
Downloads 30d
898 K
Likes
839
Commercial use: allowed apache-2.0 Not gated GGUF View on Hugging Face ↗

Download history

tracking started — chart appears after 7 days of snapshots (2 recorded)
898 K downloads in the last 30 days

Can you run it?

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

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
gemma-4-12b-it-Q3_K_S.gguf Q3_K_S 5.1 GB 7.9 GB ✅ Runs comfortably
gemma-4-12b-it-Q3_K_M.gguf Q3_K_M 5.7 GB 8.6 GB ✅ Runs comfortably
gemma-4-12b-it-IQ4_XS.gguf IQ4_XS 6.4 GB 9.3 GB ✅ Runs comfortably
gemma-4-12b-it-IQ4_NL.gguf IQ4_NL 6.7 GB 9.7 GB ✅ Runs comfortably
gemma-4-12b-it-Q4_0.gguf Q4_0 6.7 GB 9.7 GB ✅ Runs comfortably
gemma-4-12b-it-Q4_1.gguf Q4_1 7.4 GB 10.4 GB ✅ Runs comfortably
gemma-4-12b-it-Q8_0.gguf Q8_0 12.7 GB 16.2 GB ✅ Runs comfortably
gemma-4-12b-it-BF16.gguf GGUF 23.8 GB 28.5 GB ❌ Won’t fit
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Parameters
12.0 B
Licence
apache-2.0
First seen on the Hub
2026-05-29
Training datasets
undisclosed
Added to our catalog
2026-09-20
Compare with any image-text-to-text model