Model comparison
Claude 3 Opus vs Gemma 3 27B
Gemma 3 27B is the stronger model overall, scoring 30.8 to 29.5 on the Noometry Index.
Last verified . 33 shared benchmarks.
GoogleSummary
- They share 33 benchmarks with published results for both. Claude 3 Opus scores higher in 2 categories and Gemma 3 27B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 3 27B leads 25.9 to 14.8.
- The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 74% for Gemma 3 27B.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | Gemma 3 27B | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 29.5 | 30.8 |
| Released | 2024-02-29 | 2025-03-11 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.08 |
| Output $ / M tokens | — | $0.16 |
| Results tracked | 46 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3 Opus leads
Claude 3 Opus: 32.9 (#267), Gemma 3 27B: 22.5 (#334)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LiveBench Coding | 38.6% | 39.9% |
| LMArena Coding | 1264 | 1322 |
| Aider Polyglot | — | 4.9% |
| SciCode | — | 21.2% |
| WeirdML | 19.2% | — |
| BigCodeBench Instruct | 45.5% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Too close to call
Claude 3 Opus: 24.6 (#116), Gemma 3 27B: 25.1 (#110)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning Gemma 3 27B leads
Claude 3 Opus: 14.6 (#324), Gemma 3 27B: 16.7 (#301)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| Chess Puzzles | 5% | 0% |
| LiveBench Reasoning | 40.6% | 43.8% |
| LMArena Hard Prompts | 1245 | 1340 |
| DTBench | 61.6% | 52.5% |
| LiveBench Data Analysis | 57.9% | 51.5% |
| LMCA | 17% | 12.3% |
| Epoch Capabilities Index | 126.91 | 130.04 |
| LiveBench | 49.2% | 50% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 40.4% |
| CritPt | — | 0% |
| EnigmaEval | 0.8% | — |
| ForecastBench | 58.4 | — |
| WinoGrande | 88.5% | — |
Math Gemma 3 27B leads
Claude 3 Opus: 14.8 (#299), Gemma 3 27B: 25.9 (#265)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 22.5% |
| LiveBench Math | 43.6% | 55.4% |
| LMArena Math | 1273 | 1312 |
| MATH Level 5 | 37.5% | 74% |
Knowledge Gemma 3 27B leads
Claude 3 Opus: 24.5 (#267), Gemma 3 27B: 25.5 (#261)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 47.2% | 47.7% |
| Confabulations | 22.7% | 40.3% |
| LMArena Expert | 1223 | 1304 |
| SimpleQA Verified | 12.6% | — |
| Vectara Hallucination Rate | — | 7.4% |
| MMLU | 84.6% | — |
Multimodal Gemma 3 27B leads
Claude 3 Opus: 27.1 (#116), Gemma 3 27B: 32.6 (#100)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LMArena Vision | 1023 | 1164 |
| GeoBench | — | 52% |
Multilingual Gemma 3 27B leads
Claude 3 Opus: 41.4 (#207), Gemma 3 27B: 46.9 (#155)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1258 | 1334 |
| LMArena Chinese | 1248 | 1346 |
| LMArena French | 1275 | 1368 |
| LMArena German | 1258 | 1362 |
| LMArena Japanese | 1204 | 1287 |
| LMArena Korean | 1187 | 1308 |
| LMArena Russian | 1280 | 1349 |
| LMArena Spanish | 1246 | 1349 |
Instruction Following Gemma 3 27B leads
Claude 3 Opus: 64.1 (#228), Gemma 3 27B: 70.6 (#160)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LiveBench Instruction Following | 63.9% | 74.9% |
| LMArena Instruction Following | 1248 | 1321 |
Long Context Claude 3 Opus leads
Claude 3 Opus: 38.2 (#202), Gemma 3 27B: 27.6 (#293)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LMArena Longer Query | 1259 | 1333 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Gemma 3 27B leads
Claude 3 Opus: 47.2 (#213), Gemma 3 27B: 52.5 (#168)
| Benchmark | Claude 3 Opus | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1262 | 1358 |
| LMArena Creative Writing | 1235 | 1346 |
| LMArena Multi-Turn | 1275 | 1345 |
| LiveBench Language | 50.4% | 34.6% |
| Short-Story Creative Writing | — | 79.9% |
| EQ-Bench Creative Writing | — | 1266 |
Frequently asked questions
Is Claude 3 Opus better than Gemma 3 27B?
Gemma 3 27B is the stronger model overall, scoring 30.8 to 29.5 on the Noometry Index.
Is Claude 3 Opus or Gemma 3 27B better for coding?
Claude 3 Opus scores higher on coding benchmarks: 32.9 versus 22.5 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Gemma 3 27B share?
33 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Gemma 3 27B has 43.