Model comparison
Claude 3 Haiku vs Gemini 3.5 Flash Lite
Gemini 3.5 Flash Lite is the stronger model overall, scoring 41.8 to 25.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
GoogleSummary
- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Gemini 3.5 Flash Lite in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 3.5 Flash Lite leads 64.4 to 29.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 71.1% for Gemini 3.5 Flash Lite.
Side by side
| Claude 3 Haiku | Gemini 3.5 Flash Lite | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 25.7 | 41.8 |
| Released | 2024-03-07 | 2026-07-21 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 37 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 26.1 (#320), Gemini 3.5 Flash Lite: 42.1 (#96)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| WeirdML | 9.8% | 39% |
| LMArena Coding | 1199 | 1455 |
| LMArena WebDev | — | 1440 |
| SciCode | — | 41.3% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 765.27 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Gemini 3.5 Flash Lite: 28.6 (#82)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| GDP.pdf | — | 10% |
Reasoning Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 15.7 (#303), Gemini 3.5 Flash Lite: 27.8 (#95)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1441 |
| DTBench | 50.1% | 83.5% |
| LMCA | 8.8% | 37.6% |
| Epoch Capabilities Index | 118.3 | 145.12 |
| ARC-AGI-2 | — | 10.3% |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 60.4% |
| ARC-AGI-1 | — | 53.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 22% |
| Mystery Game Puzzles | — | 19% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 9.2 (#313), Gemini 3.5 Flash Lite: 25.7 (#245)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 71.1% |
| LMArena Math | 1188 | 1431 |
| FrontierMath (Tiers 1-3) | — | 26% |
| FrontierMath Tier 4 | — | 0% |
| ProofBench | — | 13% |
| MATH Level 5 | 14.9% | — |
Knowledge Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 17.3 (#280), Gemini 3.5 Flash Lite: 50.0 (#66)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| GPQA Diamond | 36.3% | 83.3% |
| LMArena Expert | 1148 | 1430 |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |
Multimodal Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 23.1 (#123), Gemini 3.5 Flash Lite: 40.3 (#38)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Vision | 950 | 1266 |
| ScienceQA | 72% | — |
Multilingual Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 36.1 (#238), Gemini 3.5 Flash Lite: 53.8 (#59)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Non-English | 1178 | 1428 |
| LMArena Chinese | 1155 | 1467 |
| LMArena French | 1194 | 1452 |
| LMArena German | 1174 | 1454 |
| LMArena Japanese | 1102 | 1428 |
| LMArena Korean | 1108 | 1406 |
| LMArena Russian | 1204 | 1443 |
| LMArena Spanish | 1165 | 1440 |
Instruction Following Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 61.5 (#242), Gemini 3.5 Flash Lite: 75.0 (#77)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Instruction Following | 1174 | 1420 |
Long Context Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 36.1 (#232), Gemini 3.5 Flash Lite: 43.8 (#80)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Longer Query | 1191 | 1435 |
Writing & Preference Gemini 3.5 Flash Lite leads
Claude 3 Haiku: 29.8 (#286), Gemini 3.5 Flash Lite: 64.4 (#55)
| Benchmark | Claude 3 Haiku | Gemini 3.5 Flash Lite |
|---|---|---|
| LMArena Text | 1195 | 1436 |
| LMArena Creative Writing | 1157 | 1419 |
| EQ-Bench Creative Writing | 717 | 1559 |
| LMArena Multi-Turn | 1190 | 1445 |
Frequently asked questions
Is Claude 3 Haiku better than Gemini 3.5 Flash Lite?
Gemini 3.5 Flash Lite is the stronger model overall, scoring 41.8 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Gemini 3.5 Flash Lite better for coding?
Gemini 3.5 Flash Lite scores higher on coding benchmarks: 42.1 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Gemini 3.5 Flash Lite share?
25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemini 3.5 Flash Lite has 38.