Model comparison
Llama 4 Maverick vs Tulu 3 (Tülu 3) 70B
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 30.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
Meta
Allen Institute for AI (Ai2)Summary
- They share 14 benchmarks with published results for both. Llama 4 Maverick scores higher in 4 categories and Tulu 3 (Tülu 3) 70B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Tulu 3 (Tülu 3) 70B leads 23.9 to 10.1.
- The biggest single-benchmark swing is MATH Level 5: 73% for Llama 4 Maverick and 42.7% for Tulu 3 (Tülu 3) 70B.
Side by side
| Llama 4 Maverick | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Meta | Allen Institute for AI (Ai2) |
| Noometry Index | 30.9 | 33.0 |
| Released | 2025-04-05 | 2024-11-21 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.19 | — |
| Output $ / M tokens | $0.65 | — |
| Results tracked | 54 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Tulu 3 (Tülu 3) 70B leads
Llama 4 Maverick: 26.6 (#324), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1302 | 1235 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), Tulu 3 (Tülu 3) 70B: —
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Llama 4 Maverick: 10.1 (#342), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1220 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| Kagi LLM Benchmark | 55.9% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| DTBench | 61.9% | — |
| LMCA | 15.9% | — |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
Math Llama 4 Maverick leads
Llama 4 Maverick: 26.0 (#262), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | 4.4% |
| LMArena Math | 1299 | 1242 |
| MATH Level 5 | 73% | 42.7% |
| Omni-MATH | 42.2% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | 67% | 46.3% |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | 65% | — |
| LMArena Expert | 1259 | — |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), Tulu 3 (Tülu 3) 70B: —
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Llama 4 Maverick leads
Llama 4 Maverick: 42.2 (#195), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1269 | 1236 |
| LMArena Chinese | 1277 | 1249 |
| LMArena Russian | 1286 | 1246 |
| LMArena French | 1259 | — |
| LMArena German | 1291 | — |
| LMArena Japanese | 1207 | — |
| LMArena Korean | 1203 | — |
| LMArena Spanish | 1293 | — |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1267 | 1233 |
| IFEval | 90.8% | — |
Long Context Tulu 3 (Tülu 3) 70B leads
Llama 4 Maverick: 31.4 (#279), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1280 | 1224 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Tulu 3 (Tülu 3) 70B leads
Llama 4 Maverick: 38.8 (#252), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Llama 4 Maverick | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1287 | 1256 |
| LMArena Creative Writing | 1267 | 1231 |
| LMArena Multi-Turn | 1289 | 1252 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than Tulu 3 (Tülu 3) 70B?
Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 33.0 to 30.9 on the Noometry Index.
Is Llama 4 Maverick or Tulu 3 (Tülu 3) 70B better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 26.6 in the Noometry coding category.
How many benchmarks do Llama 4 Maverick and Tulu 3 (Tülu 3) 70B share?
14 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.