Qwen 2.5 7B vs Mistral 7B
Architecture Comparison
SpecQwen 2.5 7BMistral 7B
TypeDENSEDENSE
Total Parameters7.6B7.3B
Active Parameters7.6B7.3B
Layers2832
Hidden Dimension3,5844,096
Attention Heads2832
KV Heads48
Context Length131,07232,768
Precision (default)BF16BF16
Memory Requirements
PrecisionQwen 2.5 7BMistral 7B
BF16 Weights15.2 GB14.6 GB
FP8 Weights7.6 GB7.3 GB
INT4 Weights3.8 GB3.6 GB
KV-Cache / Token57344 B131072 B
Activation Estimate1.00 GB1.00 GB
Minimum GPUs Needed (BF16)
H100 SXM1 GPU1 GPU
L40S1 GPU1 GPU
Quality Benchmarks
BenchmarkQwen 2.5 7BMistral 7B
Overall7056
MMLU74.262.5
HumanEval42.832.0
GSM8K82.052.2
MT-Bench79.071.0
Qwen 2.5 7B
MMLU
74.2
HumanEval
42.8
GSM8K
82.0
MT-Bench
79.0
Mistral 7B
MMLU
62.5
HumanEval
32.0
GSM8K
52.2
MT-Bench
71.0
Capabilities
FeatureQwen 2.5 7BMistral 7B
Tool Use✓ Yes✗ No
Vision✗ No✗ No
Code✓ Yes✓ Yes
Math✓ Yes✓ Yes
Reasoning✗ No✗ No
Multilingual✓ Yes✓ Yes
Structured Output✓ Yes✗ No
API Pricing Comparison
Cheapest Output (Qwen 2.5 7B)
$0.20/M
Input: $0.20/M
Cheapest Output (Mistral 7B)
$0.07/M
Input: $0.07/M
| Provider | Qwen 2.5 7B In $/M | Out $/M | Mistral 7B In $/M | Out $/M |
|---|---|---|---|---|
| deepinfra | — | — | $0.07 | $0.07 |
| together | $0.20 | $0.20 | $0.20 | $0.20 |
| fireworks | $0.20 | $0.20 | — | — |
Recommendation Summary
- ‣Qwen 2.5 7B scores higher on overall quality (70 vs 56).
- ‣Mistral 7B is cheaper per output token ($0.07/M vs $0.20/M).
- ‣Mistral 7B has a smaller memory footprint (14.6 GB vs 15.2 GB BF16), making it easier to deploy on fewer GPUs.
- ‣Qwen 2.5 7B supports a longer context window (131,072 vs 32,768 tokens).
- ‣Qwen 2.5 7B is stronger at code generation (HumanEval: 42.8 vs 32.0).
- ‣Qwen 2.5 7B is better at math reasoning (GSM8K: 82.0 vs 52.2).