bartowski
Qwen2.5 7B Instruct
Qwen2.5 7B Instruct is a 7.62B dense qwen2 model from bartowski. Configured context length is 32,768 tokens. License: apache-2.0. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters7.6B
Context32K
Attentiongqa
Licenseapache-2.0
28 layers · 4 KV heads · head dim 128 · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| F16 | 14.2 GB | Hugging Face file bytes |
| Q8_0 | 7.54 GB | Hugging Face file bytes |
| Q6_K | 11.9 GB | Hugging Face file bytes |
| Q5_K_M | 5.07 GB | Hugging Face file bytes |
| Q5_K_S | 4.95 GB | Hugging Face file bytes |
| Q4_K_M | 4.36 GB | Hugging Face file bytes |
| Q4_K_S | 4.15 GB | Hugging Face file bytes |
| IQ4_XS | 3.93 GB | Hugging Face file bytes |
| Q4_0 | 16.5 GB | Hugging Face file bytes |
| Q3_K_L | 3.81 GB | Hugging Face file bytes |
| Q3_K_M | 3.55 GB | Hugging Face file bytes |
| IQ3_M | 3.33 GB | Hugging Face file bytes |
| Q3_K_S | 3.25 GB | Hugging Face file bytes |
| Q2_K | 6.11 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Runs | Q5_K_M | 27–37 |
| GeForce RTX 3060 12GB | Runs | Q8_0 | 23–32 |
| GeForce RTX 4060 Ti 16GB | Runs great | Q8_0 | 18–25 |
| GeForce RTX 4090 | Runs | F16 | 35–48 |
| GeForce RTX 5090 | Runs great | F16 | 63–86 |
| Apple M4 Pro | Runs | F16 | 9.6–13 |
| Apple M5 | Runs | Q8_0 | 9.8–13 |
| CPU only · DDR5-5600 | Runs on CPU | F16 | 3.1–4.3 |
Smallest devices that fit it fully
- GeForce RTX 3050 6GB — Tight fit, Q4_K_M, 19–26 tok/s
- GeForce RTX 2060 6GB — Tight fit, Q4_K_M, 38–51 tok/s
- GeForce GTX 1660 Super — Tight fit, Q4_K_M, 38–51 tok/s
- GeForce GTX 1660 Ti — Tight fit, Q4_K_M, 32–44 tok/s
- Intel Arc A380 — Tight fit, Q4_K_M, 21–28 tok/s
- GeForce RTX 5060 Ti 8GB — Runs, Q5_K_M, 42–57 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (Q8_0).
llama.cpp
llama-server -hf bartowski/Qwen2.5-7B-Instruct-GGUF:Q8_0 -c 8192 -ngl 28
Ollama
ollama run qwen2.5
LM Studio
lms get bartowski/Qwen2.5-7B-Instruct-GGUF # Then load quant Q8_0