DeepSeek·released 2023-11-291 source
DeepSeek LLM 67B benchmark scores: 4 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 4 benchmarks·best result 6.39 on MATH level 5 (unverified)·0 of 4 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| MATH level 5 | 6.39accuracy (%) | unverifiedT2 | 2023-11-29 |
| OTIS Mock AIME 2024-2025 | 0.73accuracy (%) | unverifiedT2 | 2023-11-29 |
| GPQA diamond | 0accuracy (%) | unverifiedT2 | 2023-11-29 |
| Chess Puzzles | 0accuracy (%) | unverifiedT2 | 2023-11-29 |
Claims drawn from cited facts, not live model generation.
This model was developed by DeepSeek, placing its origins within a research organization focused on building capable and efficient language systems. It dates to 2023, a vintage that situates it among the earlier waves of contemporary large-scale generative models rather than the current frontier.
3 cited facts
This model is scored on 4 tracked benchmarks, yet it currently holds no top score on any of them. Its record remains unverified, meaning outside evaluation has neither confirmed nor disputed the reported figures.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 89.71 points.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
DeepSeek LLM 67B is an AI model developed by DeepSeek, released 2023-11-29. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
DeepSeek LLM 67B has recorded scores on 4 benchmarks, each shown with its evidence status.
DeepSeek LLM 67B has recorded scores on 4 benchmarks — GPQA diamond, MATH level 5, OTIS Mock AIME 2024-2025, Chess Puzzles. The full table above shows each score with its evidence status.
0 of 4 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
DeepSeek LLM 67B has 4 tracked claims: 4 unverified.