Model comparison
GPT-4.1 vs Llama 3.2 3B
GPT-4.1 is the stronger model overall, scoring 35.9 to 28.9 on the Noometry Index. Llama 3.2 3B costs 29× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
MetaSummary
- They share 15 benchmarks with published results for both. GPT-4.1 scores higher in 7 categories and Llama 3.2 3B in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 leads 57.6 to 24.7.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 54% for GPT-4.1 and 21.9% for Llama 3.2 3B.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 131K.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 | Llama 3.2 3B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 35.9 | 28.9 |
| Released | 2025-04-14 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 33K | 118K |
| Input $ / M tokens | $2 | $0.05 |
| Output $ / M tokens | $8 | $0.33 |
| Results tracked | 52 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.1 leads
GPT-4.1: 34.4 (#238), Llama 3.2 3B: 27.6 (#319)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Coding | 1391 | 1098 |
| SWE-bench Verified | 48.5% | — |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| WeirdML | 39% | — |
| BigCodeBench Instruct | — | 23.4% |
| BigCodeBench Complete | — | 28.3% |
| CadEval | 42% | — |
| ALE-Bench | 558.1 | — |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.7 (#43), Llama 3.2 3B: 20.1 (#143)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 21.9% |
| BALROG | — | 10.1% |
Reasoning Llama 3.2 3B leads
GPT-4.1: 11.7 (#339), Llama 3.2 3B: 21.0 (#228)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1384 | 1095 |
| ARC-AGI-2 | 0.4% | — |
| SimpleBench | 27% | — |
| Kagi LLM Benchmark | 52.3% | — |
| ARC-AGI-1 | 5.5% | — |
| Chess Puzzles | 6% | — |
| EnigmaEval | 2.2% | — |
| DTBench | 68.3% | — |
| LMCA | 25.6% | — |
| Epoch Capabilities Index | 136.78 | — |
| ForecastBench | 61.5 | — |
Math Llama 3.2 3B leads
GPT-4.1: 22.3 (#280), Llama 3.2 3B: 32.4 (#214)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1370 | 1126 |
| FrontierMath (Tiers 1-3) | 6% | — |
| OTIS Mock AIME 2024-2025 | 38.3% | — |
| Omni-MATH | 47.1% | — |
| MATH Level 5 | 83% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-4.1 leads
GPT-4.1: 37.1 (#160), Llama 3.2 3B: 29.7 (#235)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1364 | 1090 |
| GPQA Diamond | 66.9% | — |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | 31.1% | — |
| MMLU-Pro | 81.1% | — |
| Vectara Hallucination Rate | 5.6% | — |
| GPQA (HELM) | 65.9% | — |
Multimodal Not comparable
GPT-4.1: 38.2 (#67), Llama 3.2 3B: —
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Vision | 1211 | — |
| GeoBench | 72% | — |
Multilingual GPT-4.1 leads
GPT-4.1: 49.4 (#133), Llama 3.2 3B: 26.2 (#281)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1370 | 1019 |
| LMArena Chinese | 1382 | 1017 |
| LMArena German | 1381 | 1056 |
| LMArena Russian | 1377 | 949 |
| LMArena French | 1382 | — |
| LMArena Japanese | 1319 | — |
| LMArena Korean | 1339 | — |
| LMArena Spanish | 1376 | — |
Instruction Following GPT-4.1 leads
GPT-4.1: 71.3 (#153), Llama 3.2 3B: 56.0 (#275)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1367 | 1089 |
| IFEval | 83.8% | — |
Long Context GPT-4.1 leads
GPT-4.1: 40.0 (#163), Llama 3.2 3B: 33.4 (#261)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1385 | 1100 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.6 (#125), Llama 3.2 3B: 24.7 (#307)
| Benchmark | GPT-4.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1383 | 1110 |
| LMArena Creative Writing | 1363 | 1094 |
| EQ-Bench Creative Writing | 1420 | 595 |
| LMArena Multi-Turn | 1398 | 1105 |
| WildBench | 85.4% | — |
Frequently asked questions
Is GPT-4.1 better than Llama 3.2 3B?
GPT-4.1 is the stronger model overall, scoring 35.9 to 28.9 on the Noometry Index. Llama 3.2 3B costs 29× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 or Llama 3.2 3B?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or Llama 3.2 3B better for coding?
GPT-4.1 scores higher on coding benchmarks: 34.4 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 131K.
How many benchmarks do GPT-4.1 and Llama 3.2 3B share?
15 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and Llama 3.2 3B has 18.