Tess-4.5-27B-GGUF
Tess-4.5 is a 27-billion-parameter open-weights model for real engineering work. It reasons through every step, works with the tools and files actually in front of it, and runs on a single GPU or a laptop. On BenchLocal it matches its base, Qwen3.8-27B, with 25% fewer tokens. This repository holds the GGUF quantizations for llama.cpp, the vision projector and the multi-token-prediction draft heads.
Params
27.0 B
Context
—
Downloads 30d
952 K
Likes
65
Download history
tracking started — chart appears after 7 days of snapshots (1 recorded)952 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.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| mmproj-Tess-4.5-27B-F16.gguf | GGUF | 0.9 GB | 5.6 GB | ✅ Runs comfortably |
| Tess-4.5-27B-Q4_K_M.gguf | Q4_K_M | 16.5 GB | 22.8 GB | ⚠️ Tight — reduce context |
| Tess-4.5-27B-Q6_K.gguf | Q6_K | 22.1 GB | 28.8 GB | ❌ Won’t fit |
| Tess-4.5-27B-Q8_0.gguf | Q8_0 | 28.6 GB | 36.0 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
- 27.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-10-18
- Training datasets
- undisclosed
- Added to our catalog
- 2026-11-10
Compare with any image-text-to-text model