Model comparison
Claude 3 Haiku vs Gemini 1.5 Flash (May 2024)
Gemini 1.5 Flash (May 2024) is the stronger model overall, scoring 32.9 to 25.7 on the Noometry Index.
Last verified . 30 shared benchmarks.
GoogleSummary
- They share 30 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Gemini 1.5 Flash (May 2024) in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 1.5 Flash (May 2024) leads 48.9 to 29.8.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 61.9% for Gemini 1.5 Flash (May 2024).
Side by side
| Claude 3 Haiku | Gemini 1.5 Flash (May 2024) | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 25.7 | 32.9 |
| Released | 2024-03-07 | 2024-05-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 37 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 26.1 (#320), Gemini 1.5 Flash (May 2024): 33.9 (#233)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| WeirdML | 9.8% | 24.9% |
| BigCodeBench Instruct | 39.4% | 43.5% |
| LMArena Coding | 1199 | 1262 |
| BigCodeBench Complete | 50.1% | 55.1% |
| HumanEval+ | 68.9% | 75.6% |
| MBPP+ | 68.8% | 67.5% |
| CadEval | 12% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Gemini 1.5 Flash (May 2024): 26.3 (#97)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| BALROG | — | 14.6% |
Reasoning Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 15.7 (#303), Gemini 1.5 Flash (May 2024): 20.8 (#212)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1257 |
| DTBench | 50.1% | 53.8% |
| Epoch Capabilities Index | 118.3 | 129.37 |
| ForecastBench | 53.2 | 53.9 |
| Kagi LLM Benchmark | 34.2% | — |
| LMCA | 8.8% | — |
| PIQA | — | 87.5% |
| WinoGrande | 74.2% | — |
Math Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 9.2 (#313), Gemini 1.5 Flash (May 2024): 20.8 (#274)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 16.3% |
| LMArena Math | 1188 | 1269 |
| MATH Level 5 | 14.9% | 61.9% |
| Omni-MATH | — | 30.4% |
| FrontierMath (Feb 2025 set) | — | 0% |
| GSM8K | — | 82.4% |
Knowledge Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 17.3 (#280), Gemini 1.5 Flash (May 2024): 26.2 (#253)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| GPQA Diamond | 36.3% | 47.3% |
| LMArena Expert | 1148 | 1233 |
| MMLU | 73.8% | 77.9% |
| MMLU-Pro | — | 67.8% |
| Confabulations | 34.2% | — |
| GPQA (HELM) | — | 43.7% |
| BoolQ | — | 85.8% |
Multimodal Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 23.1 (#123), Gemini 1.5 Flash (May 2024): 35.3 (#76)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Vision | 950 | 1141 |
| Video-MME | — | 70.3% |
| GeoBench | — | 76% |
| ScienceQA | 72% | — |
Multilingual Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 36.1 (#238), Gemini 1.5 Flash (May 2024): 43.1 (#184)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Non-English | 1178 | 1278 |
| LMArena Chinese | 1155 | 1295 |
| LMArena French | 1194 | 1257 |
| LMArena German | 1174 | 1262 |
| LMArena Japanese | 1102 | 1252 |
| LMArena Korean | 1108 | 1220 |
| LMArena Russian | 1204 | 1288 |
| LMArena Spanish | 1165 | 1242 |
Instruction Following Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 61.5 (#242), Gemini 1.5 Flash (May 2024): 67.0 (#199)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Instruction Following | 1174 | 1258 |
| IFEval | — | 83.1% |
Long Context Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 36.1 (#232), Gemini 1.5 Flash (May 2024): 39.0 (#182)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Longer Query | 1191 | 1284 |
Writing & Preference Gemini 1.5 Flash (May 2024) leads
Claude 3 Haiku: 29.8 (#286), Gemini 1.5 Flash (May 2024): 48.9 (#190)
| Benchmark | Claude 3 Haiku | Gemini 1.5 Flash (May 2024) |
|---|---|---|
| LMArena Text | 1195 | 1287 |
| LMArena Creative Writing | 1157 | 1285 |
| LMArena Multi-Turn | 1190 | 1253 |
| EQ-Bench Creative Writing | 717 | — |
| WildBench | — | 79.2% |
Frequently asked questions
Is Claude 3 Haiku better than Gemini 1.5 Flash (May 2024)?
Gemini 1.5 Flash (May 2024) is the stronger model overall, scoring 32.9 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Gemini 1.5 Flash (May 2024) better for coding?
Gemini 1.5 Flash (May 2024) scores higher on coding benchmarks: 33.9 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Gemini 1.5 Flash (May 2024) share?
30 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemini 1.5 Flash (May 2024) has 42.