Only with CPU offload

Mistral 7B Instruct v0.3 at Q4_K_M needs 5.7 GB but only 5.3 GB is addressable, so about 9% of the layers would stream from system RAM at roughly 60 GB/s. Expect around 35 tokens per second — usable for batch work, painful for chat.

91% on GPU 8K context Q4_K_M · 4.1 GB Apache 2.0 Released May 2024

Old but extremely well behaved, and permissively licensed for commercial use.

Hardware requirements What hardware do I need to run Mistral 7B Instruct v0.3? →

At Q4_K_M with an 8K context: a 8 GB GPU (RTX 4060 Laptop) 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

weights 4.1 GB
Weights 4.1 GB KV cache @ 8K 1.00 GB Runtime overhead 0.6 GB Over budget 0.4 GB past 5.3 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
F16 13.5 GB 15.1 GB — ~3.5 Reference 9.8 GB over
Q8_0 7.2 GB 8.8 GB — ~8.8 −0.1% ppl 3.5 GB over
Q6_K 5.5 GB 7.1 GB — ~15 −0.4% ppl 1.8 GB over
Q5_K_M 4.8 GB 6.4 GB — ~21 −0.8% ppl 1.1 GB over
Q4_K_M 4.1 GB 5.7 GB 4K ~35 −1.9% ppl 0.4 GB over
Q3_K_M 3.3 GB 4.9 GB 11K 62 −5.4% ppl Fits
Q2_K 2.8 GB 4.4 GB 14K 72 −15% ppl Fits

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

terminal
$ llama-server \
    -hf mistralai/Mistral-7B-Instruct-v0.3:Q4_K_M \
    -c 8192 -ngl 29

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 4.1 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.
03If it falls back to CPU silently, drop the context first, then step down a quantisation.
See all models for this rig Compare with another model

Related models on a GeForce RTX 3060 Laptop

ModelTotalTok/s est.
Mistral NeMo 12B
12.2B · same family
Q4_K_M · 8.7 GB ~9.0 CPU offload
Olmo 3 7B Instruct
7.3B · similar size
Q4_K_M · 7.2 GB ~16 CPU offload
DeepSeek-R1-Distill-Qwen 7B
7.62B · similar size
Q4_K_M · 5.3 GB ~46 CPU offload
Qwen2.5-Coder 7B
7.62B · similar size
Q4_K_M · 5.3 GB ~46 CPU offload
Ling 3.0 Tiny 7.9B-A1.3B
7.9B MoE · similar size
Q4_K_M · 5.1 GB 107 Tight fit
Gemma 4 E4B
8.0B · similar size
Q4_K_M · 5.2 GB 45 Tight fit
Llama 3.1 8B Instruct
8.03B · similar size
Q4_K_M · 6.1 GB ~25 CPU offload

Every model against the GeForce RTX 3060 Laptop →

Mistral 7B Instruct v0.3 on other hardware

DeviceTotalTok/s est.
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
5.7 GB 38 Runs great
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
5.7 GB 53 Runs great
M4 · 16 GB
10.7 GB usable · 120 GB/s
5.7 GB 16 Runs great
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
5.7 GB 150 Runs great
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
5.7 GB 143 Runs great
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
5.7 GB 55 Runs great

All 114 devices, by tier →