Model comparison
Llama 4 Maverick vs o3-pro
o3-pro is the stronger model overall, scoring 42.9 to 30.9 on the Noometry Index. Llama 4 Maverick costs 115× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
MetaSummary
- They share 12 benchmarks with published results for both. Llama 4 Maverick scores higher in 1 category and o3-pro in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 31.4.
- The biggest single-benchmark swing is Aider Polyglot: 15.6% for Llama 4 Maverick and 84.9% for o3-pro.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $20 / $80 for o3-pro.
- o3-pro accepts more context: 200K tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.
Side by side
| Llama 4 Maverick | o3-pro | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 30.9 | 42.9 |
| Released | 2025-04-05 | 2025-06-10 |
| Weights | Open | Proprietary |
| Context window | 128K | 200K |
| Max output | 4K | 100K |
| Input $ / M tokens | $0.19 | $20 |
| Output $ / M tokens | $0.65 | $80 |
| Results tracked | 54 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
Llama 4 Maverick: 26.6 (#324), o3-pro: 55.5 (#24)
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| Aider Polyglot | 15.6% | 84.9% |
| WeirdML | 24.5% | 58.2% |
| SWE-bench Verified (bash only) | 21% | — |
| SciCode | 33.1% | — |
| BigCodeBench Instruct | 49.7% | — |
| LMArena Coding | 1302 | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), o3-pro: —
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning o3-pro leads
Llama 4 Maverick: 10.1 (#342), o3-pro: 23.8 (#171)
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| ARC-AGI-2 | 0% | 4.9% |
| Kagi LLM Benchmark | 55.9% | 72.1% |
| ARC-AGI-1 | 4.4% | 59.3% |
| DTBench | 61.9% | 86.9% |
| LMCA | 15.9% | 38.5% |
| Epoch Capabilities Index | 132.2 | 147.42 |
| SimpleBench | 27.7% | — |
| NYT Connections (extended) | 8% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| LMArena Hard Prompts | 1281 | — |
| ForecastBench | 57.5 | — |
Math Not comparable
Llama 4 Maverick: 26.0 (#262), o3-pro: —
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| Omni-MATH | 42.2% | — |
| LMArena Math | 1299 | — |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), o3-pro: 29.5 (#238)
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| Confabulations | 22.6% | 14.2% |
| Vectara Hallucination Rate | 8.2% | 23.3% |
| GPQA Diamond | 67% | — |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| GPQA (HELM) | 65% | — |
| LMArena Expert | 1259 | — |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), o3-pro: —
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Not comparable
Llama 4 Maverick: 42.2 (#195), o3-pro: —
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| LMArena Non-English | 1269 | — |
| LMArena Chinese | 1277 | — |
| LMArena French | 1259 | — |
| LMArena German | 1291 | — |
| LMArena Japanese | 1207 | — |
| LMArena Korean | 1203 | — |
| LMArena Russian | 1286 | — |
| LMArena Spanish | 1293 | — |
Instruction Following Not comparable
Llama 4 Maverick: 71.7 (#146), o3-pro: —
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| IFEval | 90.8% | — |
| LMArena Instruction Following | 1267 | — |
Long Context o3-pro leads
Llama 4 Maverick: 31.4 (#279), o3-pro: 72.2 (#1)
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| Fiction.LiveBench | 46.2% | 97.2% |
| LMArena Longer Query | 1280 | — |
Writing & Preference o3-pro leads
Llama 4 Maverick: 38.8 (#252), o3-pro: 57.1 (#133)
| Benchmark | Llama 4 Maverick | o3-pro |
|---|---|---|
| Short-Story Creative Writing | 62% | 84.4% |
| LMArena Text | 1287 | — |
| LMArena Creative Writing | 1267 | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
| LMArena Multi-Turn | 1289 | — |
Frequently asked questions
Is Llama 4 Maverick better than o3-pro?
o3-pro is the stronger model overall, scoring 42.9 to 30.9 on the Noometry Index. Llama 4 Maverick costs 115× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or o3-pro?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; o3-pro lists at $20 and $80.
Is Llama 4 Maverick or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
o3-pro does, with 200K tokens against 128K.
How many benchmarks do Llama 4 Maverick and o3-pro share?
12 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and o3-pro has 12.