Model comparison
GPT-5.6 Luna vs Llama 3.2 3B
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 28.9 on the Noometry Index. Llama 3.2 3B costs 3.8× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
MetaSummary
- They share 15 benchmarks with published results for both. GPT-5.6 Luna scores higher in 9 categories and Llama 3.2 3B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 32.4.
- The biggest single-benchmark swing is BALROG: 45.6% for GPT-5.6 Luna and 10.1% for Llama 3.2 3B.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $0.20 / $1.20 for GPT-5.6 Luna.
- GPT-5.6 Luna 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-5.6 Luna | Llama 3.2 3B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 54.6 | 28.9 |
| Released | 2026-07-09 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 118K |
| Input $ / M tokens | $0.20 | $0.05 |
| Output $ / M tokens | $1.20 | $0.33 |
| Results tracked | 52 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-5.6 Luna: 54.5 (#28), Llama 3.2 3B: 27.6 (#319)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Coding | 1466 | 1098 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| SciCode | 53.6% | — |
| WeirdML | 60.9% | — |
| BigCodeBench Instruct | — | 23.4% |
| BigCodeBench Complete | — | 28.3% |
| ALE-Bench | 1,667 | — |
Agentic & Tool Use GPT-5.6 Luna leads
GPT-5.6 Luna: 34.4 (#45), Llama 3.2 3B: 20.1 (#143)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| BALROG | 45.6% | 10.1% |
| APEX-Agents | 43% | — |
| Berkeley Function Calling Leaderboard | — | 21.9% |
| GDP.pdf | 22.7% | — |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), Llama 3.2 3B: 21.0 (#228)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1095 |
| ARC-AGI-2 | 59.5% | — |
| SimpleBench | 46.8% | — |
| Kagi LLM Benchmark | 49.1% | — |
| NYT Connections (extended) | 69.4% | — |
| ARC-AGI-1 | 88% | — |
| CritPt | 20.6% | — |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 21% | — |
| DTBench | 89.1% | — |
| LMCA | 48.5% | — |
| Surface Evolver Bench | 61.9% | — |
| Epoch Capabilities Index | 156.39 | — |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), Llama 3.2 3B: 32.4 (#214)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1458 | 1126 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 60% | — |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.5 (#34), Llama 3.2 3B: 29.7 (#235)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1478 | 1090 |
| GPQA Diamond | 91.6% | — |
| SimpleQA Verified | 41% | — |
Multimodal Not comparable
GPT-5.6 Luna: 42.7 (#28), Llama 3.2 3B: —
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Vision | 1258 | — |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1457 | — |
Multilingual GPT-5.6 Luna leads
GPT-5.6 Luna: 52.8 (#78), Llama 3.2 3B: 26.2 (#281)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1417 | 1019 |
| LMArena Chinese | 1470 | 1017 |
| LMArena German | 1454 | 1056 |
| LMArena Russian | 1428 | 949 |
| LMArena French | 1456 | — |
| LMArena Japanese | 1411 | — |
| LMArena Korean | 1415 | — |
| LMArena Spanish | 1448 | — |
Instruction Following GPT-5.6 Luna leads
GPT-5.6 Luna: 75.6 (#57), Llama 3.2 3B: 56.0 (#275)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1437 | 1089 |
Long Context GPT-5.6 Luna leads
GPT-5.6 Luna: 43.9 (#82), Llama 3.2 3B: 33.4 (#261)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1436 | 1100 |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), Llama 3.2 3B: 24.7 (#307)
| Benchmark | GPT-5.6 Luna | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1431 | 1110 |
| LMArena Creative Writing | 1396 | 1094 |
| EQ-Bench Creative Writing | 1829 | 595 |
| LMArena Multi-Turn | 1434 | 1105 |
| EQ-Bench 4 | 1156 | — |
Frequently asked questions
Is GPT-5.6 Luna better than Llama 3.2 3B?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 28.9 on the Noometry Index. Llama 3.2 3B costs 3.8× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.
Which is cheaper, GPT-5.6 Luna 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-5.6 Luna lists at $0.20 and $1.20.
Is GPT-5.6 Luna or Llama 3.2 3B better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 131K.
How many benchmarks do GPT-5.6 Luna and Llama 3.2 3B share?
15 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Llama 3.2 3B has 18.