Model comparison
Claude Sonnet 4 vs Llama 4 Maverick
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.9 on the Noometry Index. Llama 4 Maverick costs 20× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 49 shared benchmarks.
MetaSummary
- They share 49 benchmarks with published results for both. Claude Sonnet 4 scores higher in 8 categories and Llama 4 Maverick in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Sonnet 4 leads 57.1 to 38.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 20.6% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Claude Sonnet 4 accepts more context: 200K tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Llama 4 Maverick | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 40.8 | 30.9 |
| Released | 2025-05-22 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | 200K | 128K |
| Max output | 64K | 4K |
| Input $ / M tokens | $3 | $0.19 |
| Output $ / M tokens | $15 | $0.65 |
| Results tracked | 58 | 54 |
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), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | 21% |
| Aider Polyglot | 61.3% | 15.6% |
| SciCode | 40% | 33.1% |
| WeirdML | 46.1% | 24.5% |
| LMArena Coding | 1414 | 1302 |
| ALE-Bench | 655.35 | 172.97 |
| GSO | 4.9% | — |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
| 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), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 5.9% | 0% |
| SimpleBench | 45.5% | 27.7% |
| Kagi LLM Benchmark | 73% | 55.9% |
| ARC-AGI-1 | 40% | 4.4% |
| CritPt | 0.3% | 0% |
| EnigmaEval | 3.1% | 0.6% |
| LMArena Hard Prompts | 1372 | 1281 |
| DTBench | 77.1% | 61.9% |
| LMCA | 29% | 15.9% |
| Epoch Capabilities Index | 141.69 | 132.2 |
| ForecastBench | 60.2 | 57.5 |
| NYT Connections (extended) | — | 8% |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 20.6% |
| Omni-MATH | 60.2% | 42.2% |
| LMArena Math | 1375 | 1299 |
| MATH Level 5 | 84.4% | 73% |
| FrontierMath (Feb 2025 set) | 4.1% | 0.7% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.8 (#108), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 79.2% | 67% |
| Humanity's Last Exam | 7.8% | 5.7% |
| MMLU-Pro | 84.3% | 81% |
| Confabulations | 13.2% | 22.6% |
| Vectara Hallucination Rate | 10.3% | 8.2% |
| GPQA (HELM) | 70.6% | 65% |
| LMArena Expert | 1372 | 1259 |
Multimodal Llama 4 Maverick leads
Claude Sonnet 4: 26.2 (#121), Llama 4 Maverick: 31.6 (#105)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1191 | 1142 |
| GeoBench | 37% | 52% |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
| SpatialViz-Bench | — | 31.8% |
Multilingual Claude Sonnet 4 leads
Claude Sonnet 4: 46.7 (#156), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1333 | 1269 |
| LMArena Chinese | 1350 | 1277 |
| LMArena French | 1363 | 1259 |
| LMArena German | 1331 | 1291 |
| LMArena Japanese | 1302 | 1207 |
| LMArena Korean | 1291 | 1203 |
| LMArena Russian | 1355 | 1286 |
| LMArena Spanish | 1357 | 1293 |
Instruction Following Too close to call
Claude Sonnet 4: 71.7 (#145), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| IFEval | 84% | 90.8% |
| LMArena Instruction Following | 1376 | 1267 |
Long Context Claude Sonnet 4 leads
Claude Sonnet 4: 33.7 (#259), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| Fiction.LiveBench | 46.9% | 46.2% |
| LMArena Longer Query | 1398 | 1280 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.1 (#132), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Claude Sonnet 4 | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1351 | 1287 |
| LMArena Creative Writing | 1345 | 1267 |
| Short-Story Creative Writing | 81.4% | 62% |
| EQ-Bench Creative Writing | 1483 | 860 |
| WildBench | 83.8% | 80% |
| LMArena Multi-Turn | 1376 | 1289 |
Frequently asked questions
Is Claude Sonnet 4 better than Llama 4 Maverick?
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.9 on the Noometry Index. Llama 4 Maverick costs 20× 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 Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Llama 4 Maverick better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4 does, with 200K tokens against 128K.
How many benchmarks do Claude Sonnet 4 and Llama 4 Maverick share?
49 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Llama 4 Maverick has 54.