Model comparison
Claude 3.5 Haiku vs Gemini 2.5 Flash
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 29.2 on the Noometry Index.
Last verified . 35 shared benchmarks.
GoogleSummary
- They share 35 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and Gemini 2.5 Flash in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.5 Flash leads 39.9 to 14.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 73.1% for Gemini 2.5 Flash.
Side by side
| Claude 3.5 Haiku | Gemini 2.5 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 29.2 | 39.3 |
| Released | 2024-10-22 | 2025-04-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 49 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Flash leads
Claude 3.5 Haiku: 32.9 (#265), Gemini 2.5 Flash: 35.8 (#220)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| Aider Polyglot | 28% | 55.1% |
| WeirdML | 30.7% | 41.9% |
| LMArena Coding | 1286 | 1424 |
| SWE-bench Verified (bash only) | — | 28.7% |
| SciCode | 27.4% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 661.88 |
Agentic & Tool Use Gemini 2.5 Flash leads
Claude 3.5 Haiku: 28.0 (#95), Gemini 2.5 Flash: 30.8 (#74)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| BALROG | 19.3% | 33.5% |
| Terminal-Bench | — | 17.1% |
| Berkeley Function Calling Leaderboard | — | 56.2% |
| TheAgentCompany | — | 41.1% |
| Vending-Bench 2 | — | 548.84 |
Reasoning Too close to call
Claude 3.5 Haiku: 17.7 (#290), Gemini 2.5 Flash: 18.1 (#286)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| CritPt | 0% | 1.1% |
| LMArena Hard Prompts | 1251 | 1422 |
| DTBench | 56.7% | 76.5% |
| Epoch Capabilities Index | 127.15 | 143.03 |
| ARC-AGI-2 | — | 2.5% |
| SimpleBench | — | 41.2% |
| Kagi LLM Benchmark | — | 56.8% |
| ARC-AGI-1 | — | 33.3% |
| EnigmaEval | — | 2.7% |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 27.5% |
| ForecastBench | — | 60.6 |
| LiveBench | 43.5% | — |
Math Gemini 2.5 Flash leads
Claude 3.5 Haiku: 14.7 (#300), Gemini 2.5 Flash: 39.9 (#98)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 73.1% |
| Omni-MATH | 22.4% | 38.5% |
| LMArena Math | 1244 | 1415 |
| FrontierMath (Feb 2025 set) | 0.3% | 4.8% |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Gemini 2.5 Flash leads
Claude 3.5 Haiku: 18.7 (#281), Gemini 2.5 Flash: 36.4 (#168)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| MMLU-Pro | 60.5% | 63.9% |
| Confabulations | 36.7% | 16.8% |
| GPQA (HELM) | 36.3% | 39% |
| LMArena Expert | 1208 | 1426 |
| GPQA Diamond | 38.1% | — |
| Humanity's Last Exam | — | 12.1% |
| Vectara Hallucination Rate | — | 7.8% |
| MMLU | 74.3% | — |
Multimodal Gemini 2.5 Flash leads
Claude 3.5 Haiku: 26.8 (#117), Gemini 2.5 Flash: 41.8 (#32)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| LMArena Vision | 1092 | 1253 |
| GeoBench | 34% | 76% |
| VPCT | — | 46.2% |
| SpatialViz-Bench | — | 36.9% |
Multilingual Gemini 2.5 Flash leads
Claude 3.5 Haiku: 40.0 (#218), Gemini 2.5 Flash: 52.3 (#88)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| LMArena Non-English | 1238 | 1409 |
| LMArena Chinese | 1229 | 1450 |
| LMArena French | 1264 | 1433 |
| LMArena German | 1237 | 1418 |
| LMArena Japanese | 1175 | 1405 |
| LMArena Korean | 1173 | 1385 |
| LMArena Russian | 1253 | 1415 |
| LMArena Spanish | 1261 | 1421 |
Instruction Following Gemini 2.5 Flash leads
Claude 3.5 Haiku: 62.9 (#234), Gemini 2.5 Flash: 75.7 (#54)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| IFEval | 79.2% | 89.8% |
| LMArena Instruction Following | 1241 | 1405 |
| LiveBench Instruction Following | 61.9% | — |
Long Context Gemini 2.5 Flash leads
Claude 3.5 Haiku: 38.3 (#200), Gemini 2.5 Flash: 47.5 (#17)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| LMArena Longer Query | 1261 | 1419 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference Gemini 2.5 Flash leads
Claude 3.5 Haiku: 42.7 (#234), Gemini 2.5 Flash: 53.8 (#157)
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| LMArena Text | 1255 | 1417 |
| LMArena Creative Writing | 1233 | 1400 |
| Short-Story Creative Writing | 73.5% | 76.5% |
| EQ-Bench Creative Writing | 1146 | 1137 |
| WildBench | 76% | 81.7% |
| LMArena Multi-Turn | 1265 | 1408 |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Gemini 2.5 Flash?
Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or Gemini 2.5 Flash better for coding?
Gemini 2.5 Flash scores higher on coding benchmarks: 35.8 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Gemini 2.5 Flash share?
35 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Gemini 2.5 Flash has 54.