Run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF locally
MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF is a large instruction-tuned chat model with 12.25 billion parameters, built on the llama architecture. It has been downloaded 161,537 times.
To run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF locally at a 4,096-token context, its quantized versions need between 6.04 GB (Q2_K, lowest quality) and 24.4 GB (GGUF, highest quality) of memory, weights plus KV cache and a system margin included.
For most users the best balance is Q5_K_M, needing about 9.71 GB. That means MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF fits entirely in the VRAM of an 8 GB GPU or larger, running fully on the GPU.
All quantizations
| Quant. | Bits | Quality | Weights | KV | Total | Speed~ | Verdict |
|---|---|---|---|---|---|---|---|
| Q2_K | 3.13 | Low | 4.46 GB | 0.78 GB | 6.04 GB | 89.6 t/s | Fits in VRAM |
| Q3_K_S | 3.61 | Fair | 5.15 GB | 0.78 GB | 6.74 GB | 77.6 t/s | Fits in VRAM |
| Q3_K_M | 3.97 | Fair | 5.67 GB | 0.78 GB | 7.25 GB | 70.6 t/s | Fits in VRAM |
| Q3_K_L | 4.29 | Good | 6.11 GB | 0.78 GB | 7.69 GB | 65.5 t/s | Fits in VRAM |
| Q4_K_S | 4.65 | Good | 6.63 GB | 0.78 GB | 8.21 GB | 60.3 t/s | Fits in VRAM |
| Q4_K_M | 4.88 | Good | 6.96 GB | 0.78 GB | 8.54 GB | 57.4 t/s | Fits in VRAM |
| Q5_K_S | 5.56 | Very good | 7.93 GB | 0.78 GB | 9.51 GB | 50.4 t/s | Fits in VRAM |
| Q5_K_M | 5.7 | Very good | 8.13 GB | 0.78 GB | 9.71 GB | 49.2 t/s | Fits in VRAM |
| Q6_K | 6.57 | Excellent | 9.37 GB | 0.78 GB | 10.95 GB | 5.3 t/s | Offload |
| Q8_0 | 8.51 | Excellent | 12.13 GB | 0.78 GB | 13.71 GB | 4.1 t/s | Offload |
| GGUF | 16.01 | Excellent | 22.82 GB | 0.78 GB | 24.4 GB | 2.2 t/s | Offload |
KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.
Frequently asked questions
How much VRAM do you need to run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF?
You need about 7.69 GB of VRAM to run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF entirely on the GPU using the Q3_K_L quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.
Can I run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF on an 8 GB GPU?
Yes. With 8 GB of VRAM you can run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF fully on the GPU using Q3_K_L (about 7.69 GB).
Can I run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF on a 16 GB GPU?
Yes. With 16 GB of VRAM you can run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF fully on the GPU using Q8_0 (about 13.71 GB).
Can I run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF on a 24 GB GPU?
Yes. With 24 GB of VRAM you can run MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF fully on the GPU using Q8_0 (about 13.71 GB).
What is the best quantization for MaziyarPanahi/Mistral-Nemo-Instruct-2407-GGUF?
If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.