Yes — with 7.0 GB to spare
Ministral 3 3B at Q4_K_M fits your GeForce RTX 5070 entirely on the GPU at 8K context, at an estimated 188 tokens per second. Past 77K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.
Fully on GPU
8K context
Q4_K_M · 2.2 GB
Apache 2.0
Released Dec 2025
Vision
Edge model with a vision encoder and a 256K window. Apache 2.0.
The VRAM budget
weights 2.2 GB
Weights 2.2 GB
KV cache @ 8K 0.81 GB
Runtime overhead 0.6 GB
Free 7.0 GB of 10.6 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 7.2 GB | 8.6 GB | 27K | 57 | Reference | Long context |
| Q8_0 | 3.8 GB | 5.2 GB | 60K | 107 | −0.1% ppl | Long context |
| Q6_K | 2.9 GB | 4.4 GB | 69K | 138 | −0.4% ppl | Long context |
| Q5_K_M | 2.5 GB | 4.0 GB | 73K | 160 | −0.8% ppl | Long context |
| Q4_K_M | 2.2 GB | 3.6 GB | 77K | 188 | −1.9% ppl | Recommended |
| Q3_K_M | 1.8 GB | 3.2 GB | 81K | 232 | −5.4% ppl | Long context |
| Q2_K | 1.5 GB | 2.9 GB | 83K | 271 | −15% ppl | Long context |
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
$ ollama pull ministral-3:3b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run ministral-3:3b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
01Download is 2.2 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.
03There is room to go to 77K context on this card.