# GPT-4o mini vs GPT-5.6 Sol

> GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 25.5 on the Noometry Index. GPT-4o mini costs 30× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4o-mini-vs-gpt-5-6-sol
- Last updated: 2026-10-10
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and GPT-5.6 Sol in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Sol leads 85.6 to 10.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 100% for GPT-5.6 Sol.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
- GPT-5.6 Sol accepts more context: 1.05M tokens versus 128K.

## Snapshot

| | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 25.5 | 65.0 |
| Rank | 343 | 7 |
| Context | 128K | 1.05M |
| Input $/M | $0.15 | $4 |
| Output $/M | $0.60 | $20 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4o mini: 22.0 (#335)
- GPT-5.6 Sol: 65.1 (#7)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| WeirdML | 11.8% | 89.4% |
| LMArena Coding | 1290 | 1498 |
| DeepSWE | — | 72.7% |
| FrontierCode | — | 47.5% |
| Aider Polyglot | 3.6% | — |
| CursorBench | — | 41.7% |
| LMArena WebDev | — | 1618 |
| FrontierSWE | — | 32.2% |
| SciCode | — | 57.1% |
| GSO | — | 76.5% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| MirrorCode | — | 20% |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 2,177 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |

## Agentic & Tool Use

- GPT-4o mini: 27.5 (#101)
- GPT-5.6 Sol: 50.3 (#7)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| BALROG | 17.4% | 60% |
| APEX-Agents | — | 51.4% |
| OSWorld 2.0 | — | 27.3% |
| τ²-bench Banking | — | 46.9% |
| PostTrainBench | — | 36.2% |
| GBAEval | — | 52.6% |
| GDP.pdf | — | 30.7% |
| LMArena Search | — | 1257 |
| Vending-Bench 2 | — | 9,619 |

## Reasoning

- GPT-4o mini: 8.7 (#347)
- GPT-5.6 Sol: 74.8 (#8)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| ARC-AGI-2 | 0% | 92.5% |
| SimpleBench | 10.7% | 71.7% |
| Kagi LLM Benchmark | 28.8% | 67% |
| Chess Puzzles | 0% | 64% |
| LMArena Hard Prompts | 1267 | 1484 |
| Mystery Game Puzzles | 12% | 58% |
| DTBench | 54.4% | 96% |
| LMCA | 10.4% | 59.2% |
| Epoch Capabilities Index | 126.56 | 161.66 |
| NYT Connections (extended) | — | 93.8% |
| ARC-AGI-1 | — | 97.5% |
| CritPt | — | 32.3% |
| EnigmaEval | — | 37.1% |
| EBR-Bench | — | 44.8% |
| LiveBench Reasoning | 32.8% | — |
| LiveBench Data Analysis | 50% | — |
| Surface Evolver Bench | — | 93.1% |
| Bench to the Future 3 | — | 0.14 |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |

## Math

- GPT-4o mini: 10.4 (#314)
- GPT-5.6 Sol: 85.6 (#9)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 89.1% |
| OTIS Mock AIME 2024-2025 | 6.9% | 100% |
| LMArena Math | 1267 | 1474 |
| FrontierMath Tier 4 | — | 82.9% |
| ProofBench | — | 83% |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| MATH Level 5 | 52.6% | — |
| FrontierMath Erdős | — | 0% |
| GSM8K | 91.3% | — |

## Knowledge

- GPT-4o mini: 17.7 (#284)
- GPT-5.6 Sol: 64.3 (#18)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 37.7% | 93.5% |
| SimpleQA Verified | 8.3% | 69.7% |
| LMArena Expert | 1235 | 1516 |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| Vectara Hallucination Rate | — | 12.4% |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |

## Multimodal

- GPT-4o mini: 25.9 (#122)
- GPT-5.6 Sol: 48.6 (#9)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1066 | 1281 |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
| Blueprint-Bench 2 | — | 33.6% |
| Furniture Assembly | — | 56.7% |
| LMArena Document | — | 1483 |

## Multilingual

- GPT-4o mini: 42.0 (#199)
- GPT-5.6 Sol: 55.3 (#32)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1266 | 1452 |
| LMArena Chinese | 1265 | 1527 |
| LMArena French | 1297 | 1477 |
| LMArena German | 1272 | 1476 |
| LMArena Japanese | 1216 | 1471 |
| LMArena Korean | 1195 | 1442 |
| LMArena Russian | 1275 | 1468 |
| LMArena Spanish | 1276 | 1441 |

## Instruction Following

- GPT-4o mini: 61.9 (#239)
- GPT-5.6 Sol: 77.7 (#16)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1258 | 1482 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |

## Long Context

- GPT-4o mini: 39.1 (#186)
- GPT-5.6 Sol: 45.4 (#42)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1289 | 1480 |

## Writing & Preference

- GPT-4o mini: 39.5 (#248)
- GPT-5.6 Sol: 73.3 (#12)

| Benchmark | GPT-4o mini | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1286 | 1457 |
| LMArena Creative Writing | 1268 | 1448 |
| EQ-Bench Creative Writing | 873 | 1972 |
| LMArena Multi-Turn | 1285 | 1460 |
| Short-Story Creative Writing | 67.2% | — |
| WildBench | 79.1% | — |
| EQ-Bench 4 | — | 1250 |
| LiveBench Language | 28.6% | — |

## FAQ

### Is GPT-4o mini better than GPT-5.6 Sol?

GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 25.5 on the Noometry Index. GPT-4o mini costs 30× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.

### Which is cheaper, GPT-4o mini or GPT-5.6 Sol?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.6 Sol lists at $4 and $20.

### Is GPT-4o mini or GPT-5.6 Sol better for coding?

GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 22.0 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.6 Sol does, with 1.05M tokens against 128K.

### How many benchmarks do GPT-4o mini and GPT-5.6 Sol share?

33 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and GPT-5.6 Sol has 65.