Updated minutes ago
Codestral 22B
Mistral AI · dense · 22B parameters · 32,768 context
Quality63.0
Architecture Details
TypeDENSE
Total Parameters22B
Active Parameters22B
Layers56
Hidden Dimension6,144
Attention Heads48
KV Heads8
Head Dimension128
Vocab Size32,768
Memory Requirements
BF16 Weights
44.0 GB
FP8 Weights
22.0 GB
INT4 Weights
11.0 GB
KV-Cache per Token229376 bytes
Activation Estimate1.50 GB
Fits on (single-node)
B200 SXM BF16B100 SXM BF16GB200 NVL72 (per GPU) BF16GB300 NVL72 (per GPU) BF16H200 SXM BF16H100 SXM BF16H100 PCIe BF16H100 NVL BF16
GPU Recommendations
H20optimal
FP8 · 1 GPU · tensorrt-llm
100/100
score
Throughput
1.1K tok/s
Cost/Month
$940
Cost/M Tokens
$0.34
H100 SXMoptimal
FP8 · 1 GPU · tensorrt-llm
95/100
score
Throughput
1.1K tok/s
Cost/Month
$1794
Cost/M Tokens
$0.65
H100 PCIeoptimal
FP8 · 1 GPU · tensorrt-llm
95/100
score
Throughput
760.5 tok/s
Cost/Month
$1794
Cost/M Tokens
$0.90
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| mistral | $0.30 | $0.90 | Cheapest |
| together | $0.90 | $0.90 |
Quality Benchmarks
MMLU65.0
HumanEval58.0
GSM8K60.0
MT-Bench73.0
Capabilities
Features
✗ Tool Use✗ Vision✓ Code✓ Math✗ Reasoning✗ Multilingual✗ Structured Output
Supported Frameworks
vllmsglangtgitensorrt-llm
Supported Precisions
BF16 (default)FP8INT4