DeepSeek LLM 67B
| Released | 2023-11-29 |
|---|---|
| Price per 1M | Not listed |
| Overall score | 23.5 (early, few results) |
| Provider | DeepSeek |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 23.5 | 4 | |
| Overall: Coding | 26.7 | 1 | |
| Overall: Reasoning | 25.6 | 3 | |
| Overall: Agentic work | 26.7 | 1 | |
| Reasoning | 26.7 | 1 | |
| Math | 24.2 | 2 | |
| General | 18.1 | 1 |
Evidence
Reasoning 26.7 ±4.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 24.6% | Epoch AI | 2025-01-27 |
Math 24.2 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 6.4% | Epoch AI | 2025-01-27 |
| OTIS Mock AIME 2024-2025 | 0.8% | Epoch AI | 2026-08-28 |
General 18.1 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,185 | LMArena | 2026-11-09 |
Names matched to this model
deepseek-llm-67b, deepseek-llm-67b-base, deepseek-llm-67b-chat