Mistral 7B v0.3 VRAM: what it needs at Q4_K_M, Q8_0 and FP16
Runs in your browser — nothing you paste leaves this page. How we prove that
LLM VRAM calculator playground
KV cache is counted at FP16. Overhead is a fixed 1.5 GiB estimate for the runtime and compute buffers. A custom size borrows the architecture of the nearest preset.
Results update as you type — press Enter to run now.
GPU tiers
- 8 GiBfitsRTX 4060, RTX 3070, RX 7600
- 12 GiBfitsRTX 4070, RTX 3060 12GB, RTX 5070
- 16 GiBfitsRTX 4060 Ti 16GB, RTX 4080, RTX 5080
- 24 GiBfitsRTX 4090, RTX 3090, RX 7900 XTX
- 32 GiBfitsRTX 5090, V100 32GB
- 48 GiBfitsRTX A6000, RTX 6000 Ada, L40S
- 80 GiBfitsA100 80GB, H100 80GB
At Q4_K_M with an 8,192-token context, Mistral 7B v0.3 needs about 6.48 GiB of VRAM: 3.98 GiB of weights, 1.00 GiB of KV cache and 1.50 GiB of runtime overhead. The smallest common GPU tier that holds it is 8 GiB (RTX 4060, RTX 3070, RX 7600).
How much VRAM Mistral 7B v0.3 needs
Mistral 7B v0.3, released in May 2024, extended the vocabulary to 32,768 tokens and added function calling; it remains a light, Apache-2.0 7B for chat and agents.
With 8,192 tokens of context the estimate is 15.5 GiB at FP16, 9.43 GiB at Q8_0 and 6.48 GiB at Q4_K_M. FP16 needs a 16 GiB card, Q8_0 needs a 12 GiB card, and Q4_K_M needs an 8 GiB card. The weights are 7 billion parameters times the bits per weight of each format (16, 8.5 and 4.89), divided by eight; everything is in GiB, the unit GPU memory is sold in.
Context length and the KV cache
The KV cache stores a key and a value vector for every token in every layer. Mistral 7B v0.3 has 32 layers, 8 KV heads and a head dimension of 128, read from its Hugging Face config.json, so at FP16 each token costs 128 KiB. That is 1.00 GiB at 8,192 tokens, and 4.00 GiB at the maximum of 32,768, where the Q4_K_M total becomes 9.48 GiB and needs a 12 GiB card. The cache does not shrink with weight quantization; only a runtime option that quantizes the cache itself does.
Which quantization to pick
Q4_K_M at 6.48 GiB fits an 8 GiB card with room for a longer context; Q8_0 at 9.43 GiB needs a 12 GiB card, and FP16 at 15.5 GiB needs a 16 GiB card.
Pitfalls
Its maximum context is 32,768 tokens, a quarter of the Llama 3.1 and Qwen2.5 models, and at that length the cache is 4.00 GiB. Unlike v0.1, v0.3 does not use sliding-window attention, so the full cache really is allocated.
Mistral Nemo 12B is the bigger, 128k-context sibling; Llama 3.1 8B and Qwen2.5 7B are the same-size alternatives.
FAQ
Questions, answered.
Tap a question to expand the answer.
Can Mistral 7B v0.3 run on an 8 GB GPU at its full 32k context?
Not at Q4_K_M: 9.48 GiB at 32,768 tokens needs a 12 GiB card. At 16,384 tokens it is 7.48 GiB, which fits.
How much VRAM does Mistral 7B need at Q8_0?
9.43 GiB at 8,192 tokens, so a 12 GiB card.
More free, private DevOps tools.
The LLM VRAM Calculator is one tool in OpsCanopy — a growing canopy of browser-based validators, converters and testers that never touch a server.
Related tools
New to AI & local LLMs? Read the AI & local LLMs guide →
42 free tools, every one offline-capable — opscanopy.com works with no signup and nothing uploaded.
Related: the full LLM VRAM Calculator, the LLM Token Counter, or every AI tool.
Estimates only; the KV cache is sized at FP16 and the overhead is a fixed allowance, so always confirm against your own runtime’s report before buying hardware. OpsCanopy is free and open.