DeepSeek-V2 (MoE-236B, May 2024)
DeepSeek-V2 (MoE-236B, May 2024) (DeepSeek) was released on May 7, 2024. In the Epoch AI capability ranking (ECI), it ranks #181 out of 267 models, with 124.78. Against models of the same generation, its best result is on ARC (AI2): 92.2%, #2 of 31.
Where DeepSeek-V2 (MoE-236B, May 2024) stands out and where it falls behind, by the numbers
DeepSeek-V2 (MoE-236B, May 2024) has published results on 7 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between Aug 11, 2023 and Aug 5, 2024 that took the same test. Against the whole archive, which includes 2023 models, a recent release lands near the top of almost everything, and that says nothing.
Where it stands out
Tests where it sits in the top third of its own generation.
Where it falls behind
Tests where it sits in the bottom half of its own generation.
It is not in the bottom half of its generation on any test with 6 or more models compared.
See all 7 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| MMLU | 78.4% | #9 / 49 | #22 / 92 | 88.1% GPT-4o (May 2024) |
| ARC (AI2) | 92.2% | #2 / 31 | #4 / 65 | 95.3% DeepSeek-V3 (Mar 2025) |
| WinoGrande | 86.3% | #4 / 30 | #5 / 61 | 89.2% Llama 3.1-405B |
| HellaSwag | 87.1% | #3 / 19 | #5 / 51 | 95.3% GPT-4 (Mar 2023) |
| PIQA | 83.9% | #4 / 17 | #6 / 43 | 88.7% GPT-4o mini |
| BIG-Bench Hard | 78.8% | #4 / 19 | #7 / 41 | 89.2% Gemini 1.5 Pro (Sept 2024) |
| TriviaQA | 80.0% | #4 / 12 | #13 / 32 | 87.6% Llama 2-70B |
For each test, the model's best effort level counts, as in the ranking. Source of the scores: Epoch AI, under CC BY 4.0.
What to compare DeepSeek-V2 (MoE-236B, May 2024) with
The models from other labs closest in capability and released in the same period, at most two per lab. Click any of them to see its page.
| Model | Lab | ECI | ELO | LiveBench | Release |
|---|---|---|---|---|---|
| Claude 3 Opus | Anthropic | 126.91 | 1,264 | n/a | Feb 29, 2024 |
| GPT-4o mini | OpenAI | 126.58 | n/a | n/a | Jul 18, 2024 |
| Llama 3.1-70B | Meta AI | 125.94 | n/a | n/a | Jul 23, 2024 |
| Llama 3.2 90B | Meta AI | 125.51 | n/a | n/a | Sep 24, 2024 |
| Qwen2-72B | Alibaba | 125.28 | n/a | n/a | Jun 7, 2024 |
| DeepSeek-V2 (MoE-236B, May 2024) | DeepSeek | 124.78 | n/a | n/a | May 7, 2024 |
| Amazon Nova Pro | Amazon | 123.80 | n/a | n/a | Dec 3, 2024 |
Other models from DeepSeek
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| DeepSeek-R1 (May 2025) | May 28, 2025 | 141.27 | 1,400 | n/a | Open weights (unrestricted) |
| DeepSeek-V3 (Mar 2025) | Mar 24, 2025 | 135.93 | 1,346 | n/a | Open weights (restricted use) |
| DeepSeek-R1-Distill-Qwen-32B | Jan 20, 2025 | 137.42 | n/a | n/a | Open weights (unrestricted) |
| DeepSeek-R1-Distill-Qwen-14B | Jan 20, 2025 | 135.44 | n/a | n/a | Open weights (unrestricted) |
| DeepSeek-V2 (MoE-236B, May 2024) | May 7, 2024 | 124.78 | n/a | n/a | Open weights (restricted use) |
| DeepSeek LLM 67B | Nov 29, 2023 | 110.47 | n/a | n/a | Open weights (restricted use) |
| DeepSeek Coder 33B | Nov 2, 2023 | 96.45 | n/a | n/a | Open weights (restricted use) |
| DeepSeek Coder 6.7B | Nov 2, 2023 | 89.77 | n/a | n/a | Open weights (restricted use) |
| DeepSeek Coder 1.3B | Nov 2, 2023 | 63.88 | n/a | n/a | Open weights (restricted use) |
Have you used DeepSeek-V2 (MoE-236B, May 2024)? Rate it
Users' opinion, not measurement: the rating is kept apart from the ranking and does not move anyone's position. One vote per browser, no sign-up, and you can change your rating later.
What people ask about DeepSeek-V2 (MoE-236B, May 2024)
Where does DeepSeek-V2 (MoE-236B, May 2024) rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, DeepSeek-V2 (MoE-236B, May 2024) ranks #181 out of 267 models, with an ECI of 124.78. The ECI is a composite index that combines several hard benchmarks into a single scale. The position changes when a new model comes out, and the page is refreshed six times a day.
What is DeepSeek-V2 (MoE-236B, May 2024) best at?
Against models of the same generation (released up to 9 months before and up to 3 months after it), the best relative results of DeepSeek-V2 (MoE-236B, May 2024) are: ARC (AI2) (#2 of 31); WinoGrande (#4 of 30); HellaSwag (#3 of 19). The yardstick is its position among the models of that period that took the same test.
Where do the numbers on this page come from?
From the same three sources as the ranking, all openly licensed: Epoch AI (ECI and individual tests, CC BY 4.0), Arena (human-preference ELO, CC BY 4.0) and LiveBench (overall average, CC BY-SA 4.0). ROO3 does not measure any model. When the page shows a figure published by the maker itself, it appears in the written analysis, labeled as the maker's, and never enters the ranking.
Where each number comes from
Epoch AI, "AI Benchmarking Hub". Published at epoch.ai. Data under the CC BY 4.0 licence, reordered by roo3.co.
The Arena (LMArena) leaderboard, from the public lmarena-ai/leaderboard-dataset on Hugging Face, under the CC BY 4.0 licence. We are not affiliated with the Arena.
LiveBench, by the LiveBench team, published at livebench.ai under the CC BY-SA 4.0 license. The overall average is recomputed by roo3.co with the same formula as the website (average of the categories), and those numbers stay under CC BY-SA 4.0. We are not affiliated with LiveBench.
← See the full ranking