gemma-2-27b-it-GGUF
Filename Quant type File Size Split Description -------- ---------- --------- ----- ----------- gemma-2-27b-it-f32.gguf f32 108.91GB true Full F32 weights. gemma-2-27b-it-Q80.gguf Q80 28.94GB false Extremely high quality, generally unneeded but max available quant. gemma-2-27b-it-Q6KL.gguf Q6KL 22.63GB false Uses Q80 for embed and output weights. Very high quality, near perfect, recommended. gemma-2-27b-it-Q6K.gguf Q...
Params
27.0 B
Context
—
Downloads 30d
158 K
Likes
175
Download history
daily snapshots · 7 days158 K157 K
Oct 19Oct 21Oct 23Oct 25
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 |
|---|---|---|---|---|
| gemma-2-27b-it-IQ2_XS.gguf | IQ2_XS | 8.4 GB | 13.8 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ2_S.gguf | IQ2_S | 8.7 GB | 14.1 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ2_M.gguf | IQ2_M | 9.4 GB | 14.9 GB | ✅ Runs comfortably |
| gemma-2-27b-it-Q2_K.gguf | Q2_K | 10.4 GB | 16.0 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ3_XXS.gguf | IQ3_XXS | 10.8 GB | 16.4 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ3_XS.gguf | IQ3_XS | 11.6 GB | 17.3 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ3_M.gguf | IQ3_M | 12.5 GB | 18.3 GB | ✅ Runs comfortably |
| gemma-2-27b-it-IQ4_XS.gguf | IQ4_XS | 14.8 GB | 20.8 GB | ⚠️ Tight — reduce context |
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 run gemma-2-27b-it-gguf # pin the quantization explicitly $ ollama run gemma-2-27b-it-gguf-iq2_m
$ huggingface-cli download bartowski/gemma-2-27b-it-GGUF-GGUF \
gemma-2-27b-it-IQ2_M.gguf --local-dir .
$ llama-cli -m gemma-2-27b-it-IQ2_M.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/gemma-2-27b-it-gguf
{
"hf_id": "bartowski/gemma-2-27b-it-GGUF",
"params_b": 27.00,
"context_length": null,
"license": { "id": "gemma", "commercial": "conditional" },
"downloads_30d": 158462,
"vram_estimates": [
{ "quant": "IQ2_M", "gb": 14.9 },
{ "quant": "IQ2_S", "gb": 14.1 }
],
"updated_at": "2026-10-19T01:00:16Z"
}
Specifications
- Parameters
- 27.0 B
- Licence
- gemma
- First seen on the Hub
- 2024-06-27
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-19
Compare with any text-generation model