Model comparison
Claude 3.5 Sonnet vs Llama-3.3-70B-Instruct
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 30.6 on the Noometry Index.
Last verified . 37 shared benchmarks.
MetaSummary
- They share 37 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 7 categories and Llama-3.3-70B-Instruct in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where Claude 3.5 Sonnet leads 39.9 to 26.4.
- The biggest single-benchmark swing is LiveBench Coding: 67.1% for Claude 3.5 Sonnet and 36.6% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Sonnet | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 34.6 | 30.6 |
| Released | 2024-06-20 | 2024-12-06 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.32 |
| Results tracked | 60 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.0 (#165), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| WeirdML | 40% | 14.4% |
| BigCodeBench Instruct | 46.8% | 46.9% |
| LiveBench Coding | 67.1% | 36.6% |
| LMArena Coding | 1342 | 1268 |
| BigCodeBench Complete | 58.6% | 57.5% |
| Aider Polyglot | 51.6% | — |
| SciCode | — | 26% |
| GSO | 4.6% | — |
| CadEval | 48% | — |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.3 (#67), Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| BALROG | 32.6% | 23% |
| Berkeley Function Calling Leaderboard | — | 31.9% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| METR Time Horizons | 45.2% | — |
Reasoning Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 23.1 (#183), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| SimpleBench | 41.4% | 19.9% |
| LiveBench Reasoning | 56.7% | 50.8% |
| LMArena Hard Prompts | 1305 | 1257 |
| DTBench | 67.8% | 59.5% |
| LiveBench Data Analysis | 55% | 49.5% |
| Epoch Capabilities Index | 133.55 | 127.33 |
| ForecastBench | 60.7 | 58.6 |
| LiveBench | 59% | 50.2% |
| CritPt | — | 0% |
| EnigmaEval | 0.9% | — |
| LMCA | — | 17.5% |
Math Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 19.2 (#288), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 5.1% |
| LiveBench Math | 52.3% | 42.2% |
| LMArena Math | 1307 | 1267 |
| MATH Level 5 | 56.9% | 41.6% |
| Omni-MATH | 27.6% | — |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Llama-3.3-70B-Instruct leads
Claude 3.5 Sonnet: 28.6 (#245), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 55.3% | 47.4% |
| Confabulations | 19.9% | 22.8% |
| LMArena Expert | 1265 | 1225 |
| MMLU | 87.3% | 86.3% |
| Humanity's Last Exam | 4.1% | — |
| MMLU-Pro | 77.7% | — |
| Vectara Hallucination Rate | — | 4.1% |
| GPQA (HELM) | 56.5% | — |
Multimodal Not comparable
Claude 3.5 Sonnet: 26.5 (#120), Llama-3.3-70B-Instruct: —
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Vision | 1125 | — |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
Multilingual Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 43.2 (#185), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1283 | 1236 |
| LMArena Chinese | 1272 | 1217 |
| LMArena French | 1305 | 1281 |
| LMArena German | 1297 | 1251 |
| LMArena Japanese | 1234 | 1150 |
| LMArena Korean | 1200 | 1143 |
| LMArena Russian | 1306 | 1252 |
| LMArena Spanish | 1290 | 1270 |
Instruction Following Llama-3.3-70B-Instruct leads
Claude 3.5 Sonnet: 68.8 (#182), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| LiveBench Instruction Following | 69.3% | 82.7% |
| LMArena Instruction Following | 1297 | 1242 |
| IFEval | 85.5% | — |
Long Context Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.9 (#167), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1311 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 52.9 (#164), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | Claude 3.5 Sonnet | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1298 | 1274 |
| LMArena Creative Writing | 1292 | 1250 |
| LMArena Multi-Turn | 1326 | 1280 |
| LiveBench Language | 53.8% | 39.2% |
| Short-Story Creative Writing | 80.3% | — |
| EQ-Bench Creative Writing | 1451 | — |
| WildBench | 79.2% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than Llama-3.3-70B-Instruct?
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 30.6 on the Noometry Index.
Is Claude 3.5 Sonnet or Llama-3.3-70B-Instruct better for coding?
Claude 3.5 Sonnet scores higher on coding benchmarks: 39.0 versus 31.0 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and Llama-3.3-70B-Instruct share?
37 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and Llama-3.3-70B-Instruct has 43.