Model comparison
Llama 3.2 3B vs o4-mini
o4-mini is the stronger model overall, scoring 41.6 to 28.9 on the Noometry Index. Llama 3.2 3B costs 16× less per token, which makes it the better buy when o4-mini's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
MetaSummary
- They share 14 benchmarks with published results for both. Llama 3.2 3B scores higher in 0 categories and o4-mini in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where o4-mini leads 54.0 to 24.7.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 21.9% for Llama 3.2 3B and 53.2% for o4-mini.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- o4-mini accepts more context: 200K tokens versus 131K.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.2 3B | o4-mini | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 28.9 | 41.6 |
| Released | 2024-09-24 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 131K | 200K |
| Max output | 118K | 100K |
| Input $ / M tokens | $0.05 | $1.10 |
| Output $ / M tokens | $0.33 | $4.40 |
| Results tracked | 18 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o4-mini leads
Llama 3.2 3B: 27.6 (#319), o4-mini: 40.9 (#127)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Coding | 1098 | 1368 |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| GSO | — | 3.6% |
| WeirdML | — | 52.6% |
| BigCodeBench Instruct | 23.4% | — |
| BigCodeBench Complete | 28.3% | — |
| CadEval | — | 62% |
| ALE-Bench | — | 826.17 |
| AlgoTune | — | 1.72 |
Agentic & Tool Use o4-mini leads
Llama 3.2 3B: 20.1 (#143), o4-mini: 32.6 (#61)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | 53.2% |
| GDPval | — | 25.3% |
| BALROG | 10.1% | — |
| METR Time Horizons | — | 63.9% |
Reasoning o4-mini leads
Llama 3.2 3B: 21.0 (#228), o4-mini: 24.6 (#162)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1351 |
| ARC-AGI-2 | — | 6.1% |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| ARC-AGI-1 | — | 58.7% |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 26% |
| EnigmaEval | — | 9.2% |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 77.6% |
| LMCA | — | 26.5% |
| Epoch Capabilities Index | — | 145.64 |
| ForecastBench | — | 61.8 |
Math o4-mini leads
Llama 3.2 3B: 32.4 (#214), o4-mini: 40.8 (#89)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Math | 1126 | 1389 |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 81.7% |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge o4-mini leads
Llama 3.2 3B: 29.7 (#235), o4-mini: 43.6 (#91)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Expert | 1090 | 1343 |
| GPQA Diamond | — | 79.6% |
| Humanity's Last Exam | — | 18.1% |
| SimpleQA Verified | — | 19.6% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |
Multimodal Not comparable
Llama 3.2 3B: —, o4-mini: 40.2 (#49)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Vision | — | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
Multilingual o4-mini leads
Llama 3.2 3B: 26.2 (#281), o4-mini: 47.0 (#154)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Non-English | 1019 | 1337 |
| LMArena Chinese | 1017 | 1354 |
| LMArena German | 1056 | 1336 |
| LMArena Russian | 949 | 1334 |
| LMArena French | — | 1364 |
| LMArena Japanese | — | 1308 |
| LMArena Korean | — | 1312 |
| LMArena Spanish | — | 1347 |
Instruction Following o4-mini leads
Llama 3.2 3B: 56.0 (#275), o4-mini: 75.2 (#68)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1089 | 1321 |
| IFEval | — | 92.8% |
Long Context o4-mini leads
Llama 3.2 3B: 33.4 (#261), o4-mini: 45.5 (#33)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Longer Query | 1100 | 1315 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference o4-mini leads
Llama 3.2 3B: 24.7 (#307), o4-mini: 54.0 (#152)
| Benchmark | Llama 3.2 3B | o4-mini |
|---|---|---|
| LMArena Text | 1110 | 1353 |
| LMArena Creative Writing | 1094 | 1294 |
| LMArena Multi-Turn | 1105 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 595 | — |
| WildBench | — | 85.4% |
Frequently asked questions
Is Llama 3.2 3B better than o4-mini?
o4-mini is the stronger model overall, scoring 41.6 to 28.9 on the Noometry Index. Llama 3.2 3B costs 16× less per token, which makes it the better buy when o4-mini's lead doesn't matter for your workload.
Which is cheaper, Llama 3.2 3B or o4-mini?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; o4-mini lists at $1.10 and $4.40.
Is Llama 3.2 3B or o4-mini better for coding?
o4-mini scores higher on coding benchmarks: 40.9 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
o4-mini does, with 200K tokens against 131K.
How many benchmarks do Llama 3.2 3B and o4-mini share?
14 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and o4-mini has 60.