Codestral 22B VRAM Requirements, GGUF Sizes & GPU Guide

Interactive Codestral 22B 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: 59.2% KV: 33.1% Overhead: 7.7%

Total Required Memory

21.18 GB

Per-Device Allocation: 21.18 GB

Model Weights

12.53 GB

KV Cache (FP16): 7.00 GB

Est. Single-Stream Decode

27.9 tok/s

Runtime & TP Overhead: 1.64 GB

Max Context Before OOM

43,438

tokens

Architecture KV compression saves 83.3% vs naive MHA FP16 (42.00 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	41.35 GB	45.61 GB	50.86 GB	71.86 GB	NVIDIA A100 80GB
FP8	8.50 bpw	21.97 GB	25.64 GB	30.89 GB	51.89 GB	RTX 5090 32GB
Q8_0	8.50 bpw	21.97 GB	25.64 GB	30.89 GB	51.89 GB	RTX 5090 32GB
Q6_K	6.56 bpw	16.95 GB	20.48 GB	25.73 GB	46.73 GB	RTX 5090 32GB
Q5_K_M	5.69 bpw	14.71 GB	18.16 GB	23.41 GB	44.41 GB	RTX 5090 32GB
Q4_K_M	4.85 bpw	12.53 GB	15.93 GB	21.18 GB	42.18 GB	RTX 3090 24GB
AWQ INT4	4.50 bpw	11.63 GB	14.99 GB	20.24 GB	41.24 GB	RTX 3090 24GB
IQ4_XS	4.25 bpw	10.98 GB	14.33 GB	19.58 GB	40.58 GB	RTX 3090 24GB

Hardware Compatibility & Decode Speed (Codestral 22B @ 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	21.18 GB (88%)	Tight Fit (>90% Usable)	27.9 tok/s
RTX 4090 24GB	24 GB · 1008 GB/s	21.18 GB (88%)	Tight Fit (>90% Usable)	30.1 tok/s
RTX 5090 32GB	32 GB · 1792 GB/s	21.18 GB (66%)	Fits Comfortably	53.4 tok/s
Apple M-Max 64GB	64 GB · 546 GB/s	20.65 GB (32%)	Fits Comfortably	15.3 tok/s
Mac Studio 128GB	128 GB · 546 GB/s	20.65 GB (16%)	Fits Comfortably	15.3 tok/s
Mac Studio 192GB	192 GB · 800 GB/s	20.65 GB (11%)	Fits Comfortably	22.5 tok/s
NVIDIA A100 80GB	80 GB · 2039 GB/s	21.18 GB (27%)	Fits Comfortably	64.4 tok/s
NVIDIA H100 80GB	80 GB · 3350 GB/s	21.18 GB (27%)	Fits Comfortably	105.8 tok/s

Verified Transformer Architecture Parameters

Checkpoint ID: mistralai/Codestral-22B-v0.1
Ollama Tag: codestral:22b
License / Release: Mistral Non-Production · 2024-Q3
Topology & Attention: Dense Transformer
Attention Mechanism: Grouped-Query Attention (GQA)
Total Params (B): 22.2B
Active Params / Token (B): 22.2B
Hidden Layers / Hidden Size: 56 / 6144
Query Heads / KV Heads: 48Q / 8KV (d=128)
Vocab / Max Position: 32,768 / 32,768

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 Codestral 22B

vllm serve mistralai/Codestral-22B-v0.1 \
  --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 codestral-22b-custom -f Modelfile
FROM codestral:22b
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