Mistral Small 3.1 (24B) VRAM Requirements, GGUF Sizes & GPU G…

Interactive Mistral Small 3.1 (24B) Memory & Context Simulator

Weight Quantization (bpw)

KV Cache Precision

Target GPU / Unified Memory Tier

Devices (TP)

Context Length (tokens) 32,768 tokens

Memory Breakdown & Bandwidth Ceiling

Tight Fit (>90% Usable)

Model Weights: 67.2% KV: 24.8% Overhead: 8%

Total Required Memory

20.16 GB

Per-Device Allocation: 20.16 GB

Model Weights

13.55 GB

KV Cache (FP16): 5.00 GB

Est. Single-Stream Decode

29.4 tok/s

Runtime & TP Overhead: 1.61 GB

Max Context Before OOM

53,954

tokens

Architecture KV compression saves 75.0% vs naive MHA FP16 (20.00 GB).

Precision & Quantization VRAM Matrix (8K / 32K / 128K Context)

Total GPU memory required across weight precisions and context lengths (batch = 1, FP16 KV cache)
Precision / Format	Effective bpw	Weights Only	Total @ 8K ctx	Total @ 32K ctx	Total @ 128K ctx	Recommended Tier (@ 32K)
FP16 / BF16	16.00 bpw	44.70 GB	48.50 GB	52.25 GB	67.25 GB	NVIDIA A100 80GB
FP8	8.50 bpw	23.75 GB	26.91 GB	30.66 GB	45.66 GB	RTX 5090 32GB
Q8_0	8.50 bpw	23.75 GB	26.91 GB	30.66 GB	45.66 GB	RTX 5090 32GB
Q6_K	6.56 bpw	18.33 GB	21.33 GB	25.08 GB	40.08 GB	RTX 5090 32GB
Q5_K_M	5.69 bpw	15.90 GB	18.83 GB	22.58 GB	37.58 GB	RTX 3090 24GB
Q4_K_M	4.85 bpw	13.55 GB	16.41 GB	20.16 GB	35.16 GB	RTX 3090 24GB
AWQ INT4	4.50 bpw	12.57 GB	15.40 GB	19.15 GB	34.15 GB	RTX 3090 24GB
IQ4_XS	4.25 bpw	11.87 GB	14.68 GB	18.43 GB	33.43 GB	RTX 3090 24GB

Hardware Compatibility & Decode Speed (Mistral Small 3.1 (24B) @ Q4_K_M, 32K Context)

Fit status, memory utilization, and estimated decode throughput across flagship GPUs and Apple Silicon
Hardware Tier	Memory & Bandwidth	Required (32K)	Fit Verdict	Est. Decode Speed
RTX 3090 24GB	24 GB · 936 GB/s	20.16 GB (84%)	Tight Fit (>90% Usable)	29.4 tok/s
RTX 4090 24GB	24 GB · 1008 GB/s	20.16 GB (84%)	Tight Fit (>90% Usable)	31.7 tok/s
RTX 5090 32GB	32 GB · 1792 GB/s	20.16 GB (63%)	Fits Comfortably	56.3 tok/s
Apple M-Max 64GB	64 GB · 546 GB/s	19.62 GB (31%)	Fits Comfortably	16.1 tok/s
Mac Studio 128GB	128 GB · 546 GB/s	19.62 GB (15%)	Fits Comfortably	16.1 tok/s
Mac Studio 192GB	192 GB · 800 GB/s	19.62 GB (10%)	Fits Comfortably	23.6 tok/s
NVIDIA A100 80GB	80 GB · 2039 GB/s	20.16 GB (25%)	Fits Comfortably	67.8 tok/s
NVIDIA H100 80GB	80 GB · 3350 GB/s	20.16 GB (25%)	Fits Comfortably	111.4 tok/s

Verified Transformer Architecture Parameters

Checkpoint ID: mistralai/Mistral-Small-3.1-24B-Instruct-2503
Ollama Tag: mistral-small3.1:24b
License / Release: Apache-2.0 · 2025-Q1
Topology & Attention: Dense Transformer
Attention Mechanism: Grouped-Query Attention (GQA)
Total Params (B): 24B
Active Params / Token (B): 24B
Hidden Layers / Hidden Size: 40 / 5120
Query Heads / KV Heads: 32Q / 8KV (d=128)
Vocab / Max Position: 131,072 / 131,072

Quantized formats store per-block scale factors and zero-points alongside packed integer weights. A 4-bit GGUF Q4_K_M checkpoint averages 4.85 effective bits per parameter (not 4.00), while Q8_0 averages 8.50 bpw. For Mixture-of-Experts models, all routed and shared experts remain resident in memory to prevent token-by-token host paging.

Standard Grouped-Query Attention allocates 2 × L × H_kv × d_head × S × B × bytes_kv across layers. DeepSeek Multi-head Latent Attention (MLA) compresses keys and values into L × (d_kv_lora + d_qk_rope) × S × B × bytes_kv, slashing long-context memory by over 90%. Hybrid sliding-window models cap local-attention layers at min(S, window_size) while scaling global layers linearly with S.

Quick-Start vLLM & Ollama Commands for Mistral Small 3.1 (24B)

vllm serve mistralai/Mistral-Small-3.1-24B-Instruct-2503 \
  --tensor-parallel-size 1 \
  --max-model-len 32768 \
  --max-num-seqs 8 \
  --gpu-memory-utilization 0.95 \
  --kv-cache-dtype fp8 \
  --enable-prefix-caching \
  --port 8000

# Save as Modelfile, then run: ollama create mistral-small-24b-custom -f Modelfile
FROM mistral-small3.1:24b
PARAMETER num_ctx 32768
PARAMETER num_batch 512
PARAMETER flash_attn true

# Environment variables for ollama serve:
# OLLAMA_NUM_PARALLEL=8 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 ollama serve