Key Specifications
| Vendor | mistral |
|---|
| Version | large-2 |
|---|
| Release Date | 2024-07-24 |
|---|
| Context Window | 128000 tokens |
|---|
| Input Modalities | text |
|---|
| Output Modalities | text |
|---|
| License | Mistral Research License |
|---|
| Documentation | https://docs.mistral.ai/ |
|---|
Benchmark Performance
| Benchmark | Score | Unit | Evaluated At | Notes | Source |
|---|
| MMLU | 84 | % | 2024-07-24 | 5-shot | view |
| HUMANEVAL | 92 | pass@1 | 2024-07-24 | — | view |
| GSM8K | 93 | % | 2024-07-24 | 0-shot CoT | view |
| MATH | 71 | % | 2024-07-24 | 0-shot CoT | view |
| BBH | 81 | % | 2024-07-24 | 3-shot CoT | view |
Pricing
| Tier | Price | Currency |
|---|
| Input | $2 / Mtok | USD |
| Output | $6 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://mistral.ai/technology/
· as of 2024-08-01
Compliance
- Data Residency: EU
- SOC2: ✓
- HIPAA: ✗
- GDPR: ✓
- ISO 27001: ✓
Mistral Large 2
Ikhtisar Model
Mistral Large 2 是 Mistral AI 旗舰模型，123B 参数，128K 上下文窗口，在代码生成、数学推理与多语言任务上表现强劲，支持欧盟数据驻留与 GDPR/ISO 27001 合规。
Spesifikasi Inti
| Vendor | Versi | Tanggal Rilis | Jendela Konteks | Modalitas Input | Modalitas Output | Lisensi |
|---|
| Mistral | large-2 | 2024-07-24 | 128K | text | text | Mistral Research License |
Kinerja Benchmark
| Benchmark | Skor | Satuan | Catatan |
|---|
| MMLU (Massive Multitask Language Understanding) | 84.0 | % | 5-shot |
| HumanEval | 92.0 | pass@1 | — |
| GSM8K (Grade School Math 8K) | 93.0 | % | 0-shot CoT |
| MATH | 71.0 | % | 0-shot CoT |
| BBH (BIG-Bench Hard) | 81.0 | % | 3-shot CoT |
Harga
| Input | Output | Baca Cache | Tulis Cache |
|---|
| — | — | — | — |
per juta token
Kelebihan
- MMLU score 84.0, strong knowledge reasoning.
- HumanEval 92.0, excellent code generation.
- GSM8K 93.0, robust math reasoning.
Kekurangan
Kasus Penggunaan
Referensi