Model comparison
Claude 3 Haiku vs Gemini 3.7 Flash
Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 25.7 on the Noometry Index.
Last verified . 24 shared benchmarks.
GoogleSummary
- They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Gemini 3.7 Flash in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 3.7 Flash leads 69.5 to 9.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 97.2% for Gemini 3.7 Flash.
Side by side
| Claude 3 Haiku | Gemini 3.7 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 25.7 | 59.8 |
| Released | 2024-03-07 | 2026-08-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.75 |
| Output $ / M tokens | — | $3.75 |
| Results tracked | 37 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.7 Flash leads
Claude 3 Haiku: 26.1 (#320), Gemini 3.7 Flash: 55.7 (#19)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Coding | 1199 | 1499 |
| DeepSWE | — | 65.5% |
| FrontierCode | — | 43.6% |
| LMArena WebDev | — | 1593 |
| FrontierSWE | — | 20.3% |
| SciCode | — | 59.8% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 904.3 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Gemini 3.7 Flash: 42.2 (#18)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| APEX-Agents | — | 67.8% |
| Remote Labor Index | — | 5% |
| GDP.pdf | — | 23.8% |
Reasoning Gemini 3.7 Flash leads
Claude 3 Haiku: 15.7 (#303), Gemini 3.7 Flash: 70.1 (#12)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1497 |
| DTBench | 50.1% | 96.8% |
| LMCA | 8.8% | 50.4% |
| Epoch Capabilities Index | 118.3 | 157.72 |
| ARC-AGI-2 | — | 84.6% |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 94% |
| ARC-AGI-1 | — | 95.5% |
| CritPt | — | 14.3% |
| Chess Puzzles | — | 47% |
| Mystery Game Puzzles | — | 37% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Gemini 3.7 Flash leads
Claude 3 Haiku: 9.2 (#313), Gemini 3.7 Flash: 69.5 (#18)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 97.2% |
| LMArena Math | 1188 | 1511 |
| FrontierMath (Tiers 1-3) | — | 71.6% |
| FrontierMath Tier 4 | — | 36.6% |
| ProofBench | — | 58% |
| MATH Level 5 | 14.9% | — |
Knowledge Gemini 3.7 Flash leads
Claude 3 Haiku: 17.3 (#280), Gemini 3.7 Flash: 69.8 (#4)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| GPQA Diamond | 36.3% | 94.8% |
| LMArena Expert | 1148 | 1508 |
| SimpleQA Verified | — | 69.2% |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |
Multimodal Gemini 3.7 Flash leads
Claude 3 Haiku: 23.1 (#123), Gemini 3.7 Flash: 36.9 (#69)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Vision | 950 | 1314 |
| Furniture Assembly | — | 26.7% |
| ScienceQA | 72% | — |
Multilingual Gemini 3.7 Flash leads
Claude 3 Haiku: 36.1 (#238), Gemini 3.7 Flash: 57.8 (#7)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1178 | 1485 |
| LMArena Chinese | 1155 | 1553 |
| LMArena French | 1194 | 1512 |
| LMArena German | 1174 | 1485 |
| LMArena Japanese | 1102 | 1506 |
| LMArena Korean | 1108 | 1486 |
| LMArena Russian | 1204 | 1519 |
| LMArena Spanish | 1165 | 1507 |
Instruction Following Gemini 3.7 Flash leads
Claude 3 Haiku: 61.5 (#242), Gemini 3.7 Flash: 78.0 (#12)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1174 | 1486 |
Long Context Gemini 3.7 Flash leads
Claude 3 Haiku: 36.1 (#232), Gemini 3.7 Flash: 45.8 (#28)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Longer Query | 1191 | 1493 |
Writing & Preference Gemini 3.7 Flash leads
Claude 3 Haiku: 29.8 (#286), Gemini 3.7 Flash: 71.4 (#19)
| Benchmark | Claude 3 Haiku | Gemini 3.7 Flash |
|---|---|---|
| LMArena Text | 1195 | 1487 |
| LMArena Creative Writing | 1157 | 1490 |
| EQ-Bench Creative Writing | 717 | 1723 |
| LMArena Multi-Turn | 1190 | 1495 |
Frequently asked questions
Is Claude 3 Haiku better than Gemini 3.7 Flash?
Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Gemini 3.7 Flash better for coding?
Gemini 3.7 Flash scores higher on coding benchmarks: 55.7 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Gemini 3.7 Flash share?
24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemini 3.7 Flash has 44.