Model comparison
Claude Sonnet 4 vs Gemma 3 27B
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.8 on the Noometry Index. Gemma 3 27B costs 60× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
GoogleSummary
- They share 34 benchmarks with published results for both. Claude Sonnet 4 scores higher in 8 categories and Gemma 3 27B in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Sonnet 4 leads 43.5 to 22.5.
- The biggest single-benchmark swing is Aider Polyglot: 61.3% for Claude Sonnet 4 and 4.9% for Gemma 3 27B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Claude Sonnet 4 accepts more context: 200K tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Gemma 3 27B | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 40.8 | 30.8 |
| Released | 2025-05-22 | 2025-03-11 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 64K | 8K |
| Input $ / M tokens | $3 | $0.08 |
| Output $ / M tokens | $15 | $0.16 |
| Results tracked | 58 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 43.5 (#88), Gemma 3 27B: 22.5 (#334)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| Aider Polyglot | 61.3% | 4.9% |
| SciCode | 40% | 21.2% |
| LMArena Coding | 1414 | 1322 |
| SWE-bench Verified (bash only) | 64.9% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| LiveBench Coding | — | 39.9% |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), Gemma 3 27B: 25.1 (#110)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 22.9 (#187), Gemma 3 27B: 16.7 (#301)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| Kagi LLM Benchmark | 73% | 40.4% |
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1372 | 1340 |
| DTBench | 77.1% | 52.5% |
| LMCA | 29% | 12.3% |
| Epoch Capabilities Index | 141.69 | 130.04 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 0% |
| EnigmaEval | 3.1% | — |
| LiveBench Reasoning | — | 43.8% |
| LiveBench Data Analysis | — | 51.5% |
| ForecastBench | 60.2 | — |
| LiveBench | — | 50% |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), Gemma 3 27B: 25.9 (#265)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 22.5% |
| LMArena Math | 1375 | 1312 |
| MATH Level 5 | 84.4% | 74% |
| Omni-MATH | 60.2% | — |
| LiveBench Math | — | 55.4% |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.8 (#108), Gemma 3 27B: 25.5 (#261)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 79.2% | 47.7% |
| Confabulations | 13.2% | 40.3% |
| Vectara Hallucination Rate | 10.3% | 7.4% |
| LMArena Expert | 1372 | 1304 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Gemma 3 27B leads
Claude Sonnet 4: 26.2 (#121), Gemma 3 27B: 32.6 (#100)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| LMArena Vision | 1191 | 1164 |
| GeoBench | 37% | 52% |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Too close to call
Claude Sonnet 4: 46.7 (#156), Gemma 3 27B: 46.9 (#155)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1333 | 1334 |
| LMArena Chinese | 1350 | 1346 |
| LMArena French | 1363 | 1368 |
| LMArena German | 1331 | 1362 |
| LMArena Japanese | 1302 | 1287 |
| LMArena Korean | 1291 | 1308 |
| LMArena Russian | 1355 | 1349 |
| LMArena Spanish | 1357 | 1349 |
Instruction Following Claude Sonnet 4 leads
Claude Sonnet 4: 71.7 (#145), Gemma 3 27B: 70.6 (#160)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| LMArena Instruction Following | 1376 | 1321 |
| LiveBench Instruction Following | — | 74.9% |
| IFEval | 84% | — |
Long Context Claude Sonnet 4 leads
Claude Sonnet 4: 33.7 (#259), Gemma 3 27B: 27.6 (#293)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| Fiction.LiveBench | 46.9% | 33.3% |
| LMArena Longer Query | 1398 | 1333 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.1 (#132), Gemma 3 27B: 52.5 (#168)
| Benchmark | Claude Sonnet 4 | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1351 | 1358 |
| LMArena Creative Writing | 1345 | 1346 |
| Short-Story Creative Writing | 81.4% | 79.9% |
| EQ-Bench Creative Writing | 1483 | 1266 |
| LMArena Multi-Turn | 1376 | 1345 |
| WildBench | 83.8% | — |
| LiveBench Language | — | 34.6% |
Frequently asked questions
Is Claude Sonnet 4 better than Gemma 3 27B?
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.8 on the Noometry Index. Gemma 3 27B costs 60× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4 or Gemma 3 27B?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Gemma 3 27B better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4 does, with 200K tokens against 131K.
How many benchmarks do Claude Sonnet 4 and Gemma 3 27B share?
34 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Gemma 3 27B has 43.