Models · Compare
DeepSeek-V2.5 vs Claude 3.5 Haiku
Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.
Specs
| Field | A: DeepSeek-V2.5 | B: Claude 3.5 Haiku |
|---|---|---|
| Released | — | 2024-11-04 |
| Developer | DeepSeek | Anthropic |
| Openness | Open weights | Proprietary |
| License | DeepSeek License | Proprietary |
| OSI-approved | no | no |
| Data released | no | no |
| Training code | no | no |
| Architecture | moe | unknown |
| Total params | — | — |
| Active params | — | — |
| Experts | — | — |
| Context window | — | 200K |
| Attention | mla | unknown |
| Position enc. | rope-yarn | unknown |
| Pretraining tokens | — | — |
| Post-training | sft, rlhf | rlhf, constitutional |
| Training hardware | — | — |
| $/M input | $0.00 | $0.80 |
| $/M output | $0.00 | $4.00 |
| Output tok/sec | 0 | 0 |
Benchmarks
Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.
General reasoning
| MMLU-Pro | — | 63.4 2026-05-21 |
| GPQA-Diamond | — | 40.8 2026-05-21 |
Code
| HumanEval | 89.0 2024-09-05 | — |
| SWE-Bench Verified | — | 40.6 2024-10-22 |
| LiveCodeBench | 41.8 2024-09-05 | 31.4 2026-05-21 |
Math
| MATH | — | 72.1 2026-05-21 |
| AIME 2024 | — | 3.3 2026-05-21 |
Context · A
Intermediate checkpoint between DeepSeek-V2 and DeepSeek-V3. Merged the previously separate Chat (V2-0628) and Coder (Coder-V2-0724) tracks into a single model, reusing the V2 architecture (236B MoE, 21B active, MLA, 128K context) and adding function calling, JSON mode, and FIM completion.
Context · B
Announced October 22 2024 alongside the upgraded Claude 3.5 Sonnet and the computer-use beta; generally available in early November. Anthropic positioned it as matching Claude 3 Opus on many intelligence benchmarks at Haiku-tier speeds, with a SWE-Bench Verified score of 40.6% making it briefly the best small-tier coding model from a major lab.