Model comparison
Claude Haiku 4.5 vs DeepSeek LLM 67B
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.8 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 7 categories and DeepSeek LLM 67B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 43.0 to 8.1.
- The biggest single-benchmark swing is MATH Level 5: 96.4% for Claude Haiku 4.5 and 6.4% for DeepSeek LLM 67B.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | DeepSeek LLM 67B | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 39.3 | 24.8 |
| Released | 2025-10-15 | 2023-11-29 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $5 | — |
| Results tracked | 53 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.9 (#71), DeepSeek LLM 67B: 31.3 (#273)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| LMArena Coding | 1454 | 1096 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| WeirdML | 45.4% | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), DeepSeek LLM 67B: —
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Too close to call
Claude Haiku 4.5: 15.1 (#310), DeepSeek LLM 67B: 15.9 (#300)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| Chess Puzzles | 8% | 0% |
| LMArena Hard Prompts | 1420 | 1070 |
| Epoch Capabilities Index | 142.42 | 110.54 |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| ForecastBench | 61.4 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), DeepSeek LLM 67B: 8.1 (#318)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 0.8% |
| LMArena Math | 1395 | 1108 |
| MATH Level 5 | 96.4% | 6.4% |
| Omni-MATH | 56.1% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.6 (#146), DeepSeek LLM 67B: 7.1 (#308)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| GPQA Diamond | 71.2% | 24.6% |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
| LMArena Expert | 1443 | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), DeepSeek LLM 67B: —
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 50.1 (#124), DeepSeek LLM 67B: 29.5 (#262)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| LMArena Non-English | 1376 | 1073 |
| LMArena Chinese | 1415 | 1132 |
| LMArena French | 1408 | — |
| LMArena German | 1377 | — |
| LMArena Japanese | 1339 | — |
| LMArena Korean | 1346 | — |
| LMArena Russian | 1380 | — |
| LMArena Spanish | 1420 | — |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.6 (#144), DeepSeek LLM 67B: 55.7 (#272)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| LMArena Instruction Following | 1414 | 1079 |
| IFEval | 80.1% | — |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#86), DeepSeek LLM 67B: 33.1 (#260)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| LMArena Longer Query | 1427 | 1092 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), DeepSeek LLM 67B: 31.8 (#277)
| Benchmark | Claude Haiku 4.5 | DeepSeek LLM 67B |
|---|---|---|
| LMArena Text | 1396 | 1105 |
| LMArena Creative Writing | 1371 | 1067 |
| LMArena Multi-Turn | 1409 | 1082 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than DeepSeek LLM 67B?
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.8 on the Noometry Index.
Is Claude Haiku 4.5 or DeepSeek LLM 67B better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 31.3 in the Noometry coding category.
How many benchmarks do Claude Haiku 4.5 and DeepSeek LLM 67B share?
15 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and DeepSeek LLM 67B has 15.