Key Takeaways
Mistral Small 3 wins:
- Cheaper output tokens
Qwen2.5 7B Instruct wins:
- Cheaper input tokens
- Higher intelligence benchmark
Price Advantage
Mistral Small 3
Benchmark Advantage
Qwen2.5 7B Instruct
Context Window
Qwen2.5 7B Instruct
Speed
N/A
Pricing Comparison
Price Comparison
| Metric | Mistral Small 3 | Qwen2.5 7B Instruct | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $0.05 | $0.04 | Qwen2.5 7B Instruct |
| Output (per 1M tokens) | $0.08 | $0.10 | Mistral Small 3 |
Using a 3:1 input/output ratio, Qwen2.5 7B Instruct is 4% cheaper overall.
Mistral Small 3 Providers
DeepInfra $0.05 (Cheapest)
Chutes $0.07
Together $0.10
Qwen2.5 7B Instruct Providers
Phala $0.04 (Cheapest)
AtlasCloud $0.04 (Cheapest)
Together $0.30
Benchmark Comparison
4
Benchmarks Compared
0
Mistral Small 3 Wins
0
Qwen2.5 7B Instruct Wins
Benchmark Scores
| Benchmark | Mistral Small 3 | Qwen2.5 7B Instruct | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | - | 35.2 | - |
MMLU Pro Academic knowledge | - | 36.5 | - |
GPQA Graduate-level science | - | 5.5 | - |
BBH Big-Bench Hard | - | 34.9 | - |
Both models show comparable benchmark performance.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
Other models
Context & Performance
Context Window
Mistral Small 3
32,768
tokens
Max output: 16,384 tokens
Qwen2.5 7B Instruct
32,768
tokens
Speed Performance
Speed benchmarks not available for these models
Capabilities
Feature Comparison
| Feature | Mistral Small 3 | Qwen2.5 7B Instruct |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | Mistral Small 3 | Qwen2.5 7B Instruct |
|---|---|---|
| License | Open Source | Open Source |
| Author | Mistral AI | Qwen |
| Released | Jan 2025 | Oct 2024 |
Mistral Small 3 Modalities
Input
text
Output
text
Qwen2.5 7B Instruct Modalities
Input
text
Output
text