Mistral Large 2 vs Mistral Large: Benchmark Comparison
Detailed comparison of Mistral Large 2 and Mistral Large covering benchmarks, pricing, context window, and compliance.
Key Specifications
| Specification | Mistral Large 2 | Mistral Large |
|---|---|---|
| Vendor | mistral | mistral |
| Version | large-2 | large |
| Release Date | 2024-07-24 | 2024-02-26 |
| Context Window | 128000 tokens | 32000 tokens |
| Input Modalities | text | text |
| Output Modalities | text | text |
| License | Mistral Research License | Apache 2.0 |
| SOC2 | ✓ | ✓ |
| HIPAA | ✗ | ✗ |
| GDPR | ✓ | ✓ |
| ISO 27001 | ✓ | ✓ |
Benchmark Results
| Benchmark | Mistral Large 2 | Mistral Large | Winner |
|---|---|---|---|
| ARC | — | 93.3 | Mistral Large |
| BBH | 81 | 76.7 | Mistral Large 2 |
| GPQA | — | 38.1 | Mistral Large |
| GSM8K | 93 | 80.9 | Mistral Large 2 |
| HUMANEVAL | 92 | 73.4 | Mistral Large 2 |
| IFEVAL | — | 75.8 | Mistral Large |
| MATH | 71 | 49.3 | Mistral Large 2 |
| MMLU | 84 | 82 | Mistral Large 2 |
| MUSR | — | 52.1 | Mistral Large |
| WINOGRANDE | — | 83.3 | Mistral Large |
Pricing Comparison
| Tier (per Mtok) | Mistral Large 2 | Mistral Large |
|---|---|---|
| Input | $2 | $4 |
| Output | $6 | $12 |
| Cache Read | $0 | $0 |
| Cache Write | $0 | $0 |
Mistral Large 2 vs Mistral Large
Model Overview
Mistral Large 2 and Mistral Large are both notable options in the AI model market. This page compares their benchmarks, pricing, and compliance.
Key Specifications
| Vendor | Release Date | Context Window | License |
|---|---|---|---|
| Mistral / Mistral | 2024-07-24 / 2024-02-26 | 128K / 32K | Mistral Research License / Apache 2.0 |
Benchmark Performance
| Benchmark | Mistral Large 2 | Mistral Large | Winner |
|---|---|---|---|
| ARC | — | 93.3 | B |
| BBH (BIG-Bench Hard) | 81.0 | 76.7 | A |
| GPQA | — | 38.1 | B |
| GSM8K (Grade School Math 8K) | 93.0 | 80.9 | A |
| HumanEval | 92.0 | 73.4 | A |
| IFEval | — | 75.8 | B |
| MATH | 71.0 | 49.3 | A |
| MMLU (Massive Multitask Language Understanding) | 84.0 | 82.0 | A |
| MUSR | — | 52.1 | B |
| WinoGrande | — | 83.3 | B |
Pricing Comparison
| Input | Output | Cache Read | Cache Write |
|---|---|---|---|
| — / — | — / — | — / — | — / — |
per million tokens — A / B
Strengths & Weaknesses
Mistral Large 2
- ✅ MMLU score 84.0, strong knowledge reasoning.
- ✅ HumanEval 92.0, excellent code generation.
- ✅ GSM8K 93.0, robust math reasoning.
- ⚠️ Proprietary, not self-hostable.
Mistral Large
- ✅ MMLU score 82.0, strong knowledge reasoning.
- ⚠️ Proprietary, not self-hostable.
Editor’s Take
Mistral Large 2 and Mistral Large each have their strengths. Choose based on workload (code, long context, vision), referencing the tables above.
FAQ
Which model is better for coding tasks?
Refer to the HumanEval benchmark table; the model with a higher score is better suited for coding tasks.
Which model is cheaper?
Refer to the pricing comparison table above; the model with lower input/output prices is more cost-effective.
Which has a longer context window?
Refer to the key specifications table; the model with a larger context window is better for long documents.
References
Editor's Take
See Editor's Take section.