Tiel-Coder-35B-A3B-GGUF-MTP
Tiel is the fast coder of the arsenal. At 4-bit quantization and 23 GB it fixes real codebase issues at the rate (and speed, with the right GPU) of Opus 4.6 medium, while holding the best multi-turn conversation of any local model we have measured. It is also cheerfully bad at trivia.
Params
35.0 B
Context
—
Downloads 30d
916 K
Likes
184
Download history
tracking started — chart appears after 7 days of snapshots (2 recorded)916 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 |
|---|---|---|---|---|
| Tiel-Coder-35B-A3B-MTP-UD-Q2_K_XL.gguf | Q2_K_XL | 12.7 GB | 19.7 GB | ⚠️ Tight — reduce context |
| Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS.gguf | IQ3_XXS | 13.6 GB | 20.7 GB | ⚠️ Tight — reduce context |
| Tiel-Coder-35B-A3B-MTP-UD-Q3_K_XL.gguf | Q3_K_XL | 17.2 GB | 24.7 GB | ❌ Won’t fit |
| Tiel-Coder-35B-A3B-MTP-UD-IQ4_XS.gguf | IQ4_XS | 18.1 GB | 25.7 GB | ❌ Won’t fit |
| Tiel-Coder-35B-A3B-MTP-UD-Q4_K_S.gguf | Q4_K_S | 21.3 GB | 29.2 GB | ❌ Won’t fit |
| Tiel-Coder-35B-A3B-MTP-UD-Q4_K_XL.gguf | Q4_K_XL | 22.7 GB | 30.8 GB | ❌ Won’t fit |
| Tiel-Coder-35B-A3B-MTP-UD-Q5_K_XL.gguf | Q5_K_XL | 27.0 GB | 35.4 GB | ❌ Won’t fit |
| Tiel-Coder-35B-A3B-MTP-UD-Q6_K_XL.gguf | Q6_K_XL | 32.2 GB | 41.2 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
- 35.0 B
- Licence
- mit
- First seen on the Hub
- 2026-08-24
- Base model
- Ornith-1.5-35B-A3B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-09-20
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any image-text-to-text model