Model comparison
GPT-5.6 Luna vs Phi 3 Small 8k Instruct
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 29.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. GPT-5.6 Luna scores higher in 8 categories and Phi 3 Small 8k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 27.6.
- Phi 3 Small 8k Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Luna | Phi 3 Small 8k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 54.6 | 29.3 |
| Released | 2026-07-09 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 52 | 32 |
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), Phi 3 Small 8k Instruct: 27.9 (#318)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Coding | 1466 | 1101 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| SciCode | 53.6% | — |
| WeirdML | 60.9% | — |
| LiveBench Coding | — | 20.3% |
| ALE-Bench | 1,667 | — |
Agentic & Tool Use Not comparable
GPT-5.6 Luna: 34.4 (#45), Phi 3 Small 8k Instruct: —
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| APEX-Agents | 43% | — |
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), Phi 3 Small 8k Instruct: 14.8 (#323)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1100 |
| 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% | — |
| LiveBench Reasoning | — | 15.9% |
| Mystery Game Puzzles | 21% | — |
| DTBench | 89.1% | — |
| LiveBench Data Analysis | — | 30.3% |
| LMCA | 48.5% | — |
| Surface Evolver Bench | 61.9% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 79.1% |
| Epoch Capabilities Index | 156.39 | — |
| HellaSwag | — | 77% |
| LiveBench | — | 24% |
| WinoGrande | — | 81.5% |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), Phi 3 Small 8k Instruct: 27.6 (#248)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Math | 1458 | 1151 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 60% | — |
| LiveBench Math | — | 17.6% |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.5 (#34), Phi 3 Small 8k Instruct: 29.1 (#240)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1478 | 1067 |
| GPQA Diamond | 91.6% | — |
| SimpleQA Verified | 41% | — |
| ARC (AI2) Challenge | — | 90.7% |
| MMLU | — | 75.7% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 58.1% |
Multimodal Not comparable
GPT-5.6 Luna: 42.7 (#28), Phi 3 Small 8k Instruct: —
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| 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), Phi 3 Small 8k Instruct: 28.5 (#272)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1417 | 1058 |
| LMArena Chinese | 1470 | 1061 |
| LMArena French | 1456 | 1135 |
| LMArena German | 1454 | 1080 |
| LMArena Japanese | 1411 | 966 |
| LMArena Korean | 1415 | 894 |
| LMArena Russian | 1428 | 1111 |
| LMArena Spanish | 1448 | 1111 |
Instruction Following GPT-5.6 Luna leads
GPT-5.6 Luna: 75.6 (#57), Phi 3 Small 8k Instruct: 51.9 (#292)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Instruction Following | 1437 | 1087 |
| LiveBench Instruction Following | — | 47.2% |
Long Context GPT-5.6 Luna leads
GPT-5.6 Luna: 43.9 (#82), Phi 3 Small 8k Instruct: 33.0 (#267)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1436 | 1088 |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), Phi 3 Small 8k Instruct: 31.1 (#284)
| Benchmark | GPT-5.6 Luna | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1431 | 1110 |
| LMArena Creative Writing | 1396 | 1083 |
| LMArena Multi-Turn | 1434 | 1068 |
| EQ-Bench Creative Writing | 1829 | — |
| EQ-Bench 4 | 1156 | — |
| LiveBench Language | — | 12.9% |
Frequently asked questions
Is GPT-5.6 Luna better than Phi 3 Small 8k Instruct?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 29.3 on the Noometry Index.
Is GPT-5.6 Luna or Phi 3 Small 8k Instruct better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 27.9 in the Noometry coding category.
How many benchmarks do GPT-5.6 Luna and Phi 3 Small 8k Instruct share?
17 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Phi 3 Small 8k Instruct has 32.