Only with CPU offload
Mistral NeMo 12B at Q4_K_M needs 8.7 GB but only 5.3 GB is addressable, so about 50% of the layers would stream from system RAM at roughly 60 GB/s. Expect around 9.0 tokens per second — usable for batch work, painful for chat.
Multilingual 12B with a 128K window, built with NVIDIA. A roleplay and fiction favourite that refuses to die.
Hardware requirements What hardware do I need to run Mistral NeMo 12B? →At Q4_K_M with an 8K context: a 12 GB GPU (RTX 3060 12 GB) or a 16 GB Mac (M4). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.
The VRAM budget
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 22.7 GB | 24.6 GB | — | ~1.8 | Reference | 19.3 GB over |
| Q8_0 | 12.1 GB | 13.9 GB | — | ~3.9 | −0.1% ppl | 8.6 GB over |
| Q6_K | 9.3 GB | 11.2 GB | — | ~5.6 | −0.4% ppl | 5.9 GB over |
| Q5_K_M | 8.1 GB | 9.9 GB | — | ~7.0 | −0.8% ppl | 4.6 GB over |
| Q4_K_M | 6.9 GB | 8.7 GB | — | ~9.0 | −1.9% ppl | 3.4 GB over |
| Q3_K_M | 5.6 GB | 7.4 GB | — | ~13 | −5.4% ppl | 2.1 GB over |
| Q2_K | 4.8 GB | 6.6 GB | — | ~19 | −15% ppl | 1.3 GB over |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.
How to run it
$ llama-server \
-hf mistralai/Mistral-Nemo-Instruct-2407:Q4_K_M \
-c 8192 -ngl 20
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
Related models on a GeForce GTX 1660 Super
| Model | Total | Tok/s est. | |
|---|---|---|---|
| Mistral 7B Instruct v0.3 7.25B · same family |
Q4_K_M · 5.7 GB | ~35 | CPU offload |
| Gemma 3 12B 12.2B · similar size |
Q4_K_M · 8.3 GB | ~9.9 | CPU offload |
| Gemma 4 12B 12B · similar size |
Q4_K_M · 8.2 GB | ~10 | CPU offload |
| Ministral 3 14B 13.9B · similar size |
Q4_K_M · 9.7 GB | ~7.3 | CPU offload |
| Phi-4 14B 14.7B · similar size |
Q4_K_M · 10.4 GB | ~6.4 | CPU offload |
| Qwen3 14B 14.8B · similar size |
Q4_K_M · 10.2 GB | ~6.6 | CPU offload |
| DeepSeek-R1-Distill-Qwen 14B 14.8B · similar size |
Q4_K_M · 10.4 GB | ~6.4 | CPU offload |
Mistral NeMo 12B on other hardware
| Device | Total | Tok/s est. | |
|---|---|---|---|
| GeForce RTX 4060 Laptop 7.0 GB usable · 256 GB/s |
8.7 GB | ~12 | CPU offload |
| GeForce RTX 3060 12 GB 10.6 GB usable · 360 GB/s |
8.7 GB | 32 | Runs great |
| M4 · 16 GB 10.7 GB usable · 120 GB/s |
8.7 GB | 9.8 | Runs great |
| GeForce RTX 4090 22.4 GB usable · 1008 GB/s |
8.7 GB | 89 | Runs great |
| Radeon RX 7900 XTX 22.4 GB usable · 960 GB/s |
8.7 GB | 85 | Runs great |
| M3 Max · 64 GB 48.0 GB usable · 400 GB/s |
8.7 GB | 33 | Runs great |