Up to four models
DeepSeek V4.1 Flash 1017 vs Claude Haiku 5.5
On the index (GPT-4 = 100), DeepSeek V4.1 Flash 1017 scores 176 and Claude Haiku 5.5 174. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
DeepSeek V4.1 Flash 1017176160–192
WeightsOpen
ReleasedOct 2026
Price in / out$0.30 / $1.2
Context1.05M
Core tests taken5 of 9
Claude Haiku 5.5
174158–191
WeightsClosed
ReleasedOct 2026
Price in / out$0.10 / $0.50
Context1M
Core tests taken7 of 9
DeepSeek V4.1 Flash 1017 and Claude Haiku 5.5 are too close to call (2 points apart, within the uncertainty), ahead on 3 of 5 shared core tests, at 2.6× the price.
Test by testthird-party results
DeepSeek V4.1 Flash 1017Claude Haiku 5.5Every result as a table
| Test | DeepSeek V4.1 Flash 1017 | Claude Haiku 5.5 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 93.2% | 89.6% |
| SimpleQA Verified | 42.5% | 23.8% |
| Maths | ||
| FrontierMath (tiers 1-3) | 71.2% | 75.1% |
| AIME-style maths (OTIS mock) | 99.4% | 98.9% |
| Chess puzzles | 46.0% | 55.0% |
| FrontierMath (tier 4) | 46.3% | 46.3% |
| Coding | ||
| Terminal-Bench 4.0 | 31.8% | 36.4% |
| Code Migration | 44.2% | 44.6% |
| Vibe Code Bench v1.1 | 87.7% | 90.4% |
| Agents | ||
| Mystery game puzzles | 49.0% | 30.0% |
| Professional work | ||
| Finance Agent (v2) | 57.0% | 54.1% |
| EMB | 61.8% | 59.0% |
| HLAB-Vals-1.1 | 14.2% | 4.2% |
| Legal Research Bench | 45.2% | 43.8% |
| Tax Agent Bench | 44.4% | 31.9% |
| Novel problems | ||
| ARC-AGI-2 | 78.1% | 61.3% |
| ARC-AGI-1 | 94.5% | 89.0% |
Tests at least two of these models have taken. Bold rows are the core tests.
