Model comparison
Claude 3.5 Haiku vs Gemma 3 27B
Gemma 3 27B is the stronger model overall, scoring 30.8 to 29.2 on the Noometry Index.
Last verified . 37 shared benchmarks.
GoogleSummary
- They share 37 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 4 categories and Gemma 3 27B in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 3 27B leads 25.9 to 14.7.
- The biggest single-benchmark swing is MATH Level 5: 46.4% for Claude 3.5 Haiku and 74% for Gemma 3 27B.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | Gemma 3 27B | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 29.2 | 30.8 |
| Released | 2024-10-22 | 2025-03-11 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.08 |
| Output $ / M tokens | — | $0.16 |
| Results tracked | 49 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.5 Haiku leads
Claude 3.5 Haiku: 32.9 (#265), Gemma 3 27B: 22.5 (#334)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| Aider Polyglot | 28% | 4.9% |
| SciCode | 27.4% | 21.2% |
| LiveBench Coding | 51.4% | 39.9% |
| LMArena Coding | 1286 | 1322 |
| WeirdML | 30.7% | — |
| BigCodeBench Instruct | 46.1% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
Agentic & Tool Use Claude 3.5 Haiku leads
Claude 3.5 Haiku: 28.0 (#95), Gemma 3 27B: 25.1 (#110)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |
| BALROG | 19.3% | — |
Reasoning Too close to call
Claude 3.5 Haiku: 17.7 (#290), Gemma 3 27B: 16.7 (#301)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| CritPt | 0% | 0% |
| LiveBench Reasoning | 28.1% | 43.8% |
| LMArena Hard Prompts | 1251 | 1340 |
| DTBench | 56.7% | 52.5% |
| LiveBench Data Analysis | 48.5% | 51.5% |
| Epoch Capabilities Index | 127.15 | 130.04 |
| LiveBench | 43.5% | 50% |
| Kagi LLM Benchmark | — | 40.4% |
| Chess Puzzles | — | 0% |
| LMCA | — | 12.3% |
Math Gemma 3 27B leads
Claude 3.5 Haiku: 14.7 (#300), Gemma 3 27B: 25.9 (#265)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 22.5% |
| LiveBench Math | 35.5% | 55.4% |
| LMArena Math | 1244 | 1312 |
| MATH Level 5 | 46.4% | 74% |
| Omni-MATH | 22.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Gemma 3 27B leads
Claude 3.5 Haiku: 18.7 (#281), Gemma 3 27B: 25.5 (#261)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 38.1% | 47.7% |
| Confabulations | 36.7% | 40.3% |
| LMArena Expert | 1208 | 1304 |
| MMLU-Pro | 60.5% | — |
| Vectara Hallucination Rate | — | 7.4% |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal Gemma 3 27B leads
Claude 3.5 Haiku: 26.8 (#117), Gemma 3 27B: 32.6 (#100)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Vision | 1092 | 1164 |
| GeoBench | 34% | 52% |
Multilingual Gemma 3 27B leads
Claude 3.5 Haiku: 40.0 (#218), Gemma 3 27B: 46.9 (#155)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1238 | 1334 |
| LMArena Chinese | 1229 | 1346 |
| LMArena French | 1264 | 1368 |
| LMArena German | 1237 | 1362 |
| LMArena Japanese | 1175 | 1287 |
| LMArena Korean | 1173 | 1308 |
| LMArena Russian | 1253 | 1349 |
| LMArena Spanish | 1261 | 1349 |
Instruction Following Gemma 3 27B leads
Claude 3.5 Haiku: 62.9 (#234), Gemma 3 27B: 70.6 (#160)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| LiveBench Instruction Following | 61.9% | 74.9% |
| LMArena Instruction Following | 1241 | 1321 |
| IFEval | 79.2% | — |
Long Context Claude 3.5 Haiku leads
Claude 3.5 Haiku: 38.3 (#200), Gemma 3 27B: 27.6 (#293)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Longer Query | 1261 | 1333 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Gemma 3 27B leads
Claude 3.5 Haiku: 42.7 (#234), Gemma 3 27B: 52.5 (#168)
| Benchmark | Claude 3.5 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1255 | 1358 |
| LMArena Creative Writing | 1233 | 1346 |
| Short-Story Creative Writing | 73.5% | 79.9% |
| EQ-Bench Creative Writing | 1146 | 1266 |
| LMArena Multi-Turn | 1265 | 1345 |
| LiveBench Language | 35.4% | 34.6% |
| WildBench | 76% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Gemma 3 27B?
Gemma 3 27B is the stronger model overall, scoring 30.8 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or Gemma 3 27B better for coding?
Claude 3.5 Haiku scores higher on coding benchmarks: 32.9 versus 22.5 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Gemma 3 27B share?
37 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Gemma 3 27B has 43.