# GPT-5.6 Sol vs GPT-5.6 Terra

> Side-by-side benchmark comparison for GPT-5.6 Sol and GPT-5.6 Terra across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-6-sol-vs-gpt-5-6-terra
- Last updated: September 24, 2026

- Shared sourced benchmarks: 32
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Pick GPT-5.6 Sol if you want the stronger benchmark profile. GPT-5.6 Terra only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.6 Sol leads overall 78.49 to 72.58.
- The clearest category separation is in multimodal & grounded, where the averages are 87.6 for GPT-5.6 Sol and 77.2 for GPT-5.6 Terra.
- The biggest single benchmark swing is FrontierMath v2 (Tier 4) in Mathematics, with scores of 83.000% and 68.300%.
- GPT-5.6 Terra is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | GPT-5.6 Sol | GPT-5.6 Terra |
|----------|----------|----------|
| Creator | OpenAI | OpenAI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 1.05M | 1.05M |
| Overall Score | 78.49 | 72.58 |
| Benchmarks Covered | 37 | 32 |
| Pricing (input/output) | $4.00 / $20.00 | $2.00 / $12.00 |

## Category Breakdown

### Agentic

- Winner: GPT-5.6 Sol
- GPT-5.6 Sol public-lane score: 69.6 (Supported · #7/105)
- GPT-5.6 Terra public-lane score: 59.5 (Supported · #18/105)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 3.0 | 34.6% | 20.8% | GPT-5.6 Sol |
| Terminal-Bench 2.1 | 91.9% | 87.4% | GPT-5.6 Sol |
| BrowseComp | 92.2% | 87.5% | GPT-5.6 Sol |
| OSWorld 2.0 | 62.6% | 50.2% | GPT-5.6 Sol |
| CyberGym | 84.5% | 81.8% | GPT-5.6 Sol |
| ExploitGym | 33.7% | 23.2% | GPT-5.6 Sol |
| Toolathlon | 58% | 53.1% | GPT-5.6 Sol |
| Terminal-Bench 2.1 (Vals) | 85.8% | 77.5% | GPT-5.6 Sol |
| ApprenticeBench | 26% | 16% | GPT-5.6 Sol |

### Coding

- Winner: GPT-5.6 Sol
- GPT-5.6 Sol public-lane score: 71.6 (Supported · #6/135)
- GPT-5.6 Terra public-lane score: 64.7 (Supported · #8/135)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| Bug Hunt Bench | 42 fixes | Coming soon | Coming soon |
| SWE-bench Pro | 64.6% | 63.4% | GPT-5.6 Sol |
| Terminal-Bench 2.1 | 91.9% | 87.4% | GPT-5.6 Sol |
| DeepSWE | 72.7% | 69.6% | GPT-5.6 Sol |
| FrontierCode 1.1 Extended | 60.6% | 55.8% | GPT-5.6 Sol |
| FrontierSWE v2 | 32.2% | Coming soon | Coming soon |
| cursorBench32 | 67.2% | 64.9% | GPT-5.6 Sol |
| VulcanBench v3 | 87.0% | 87.0% | Tie |
| VulcanBench CII v1 | 86.5% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 82.6% | 85.9% | GPT-5.6 Terra |
| SWE-bench (Vals) | 96.2% | 95.4% | GPT-5.6 Sol |
| cursorBench40 | 41.7% | Coming soon | Coming soon |

### Reasoning

- Winner: GPT-5.6 Sol
- GPT-5.6 Sol public-lane score: 72.1 (#8/19)
- GPT-5.6 Terra public-lane score: 65.4 (#12/19)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 92.5% | 83.9% | GPT-5.6 Sol |
| ARC-AGI-3 | 7.8% | 0.8% | GPT-5.6 Sol |
| GeneBench-Pro | 28.7% | Coming soon | Coming soon |

### Multimodal & Grounded

- Winner: GPT-5.6 Sol
- GPT-5.6 Sol public-lane score: 87.6 (#5/50)
- GPT-5.6 Terra public-lane score: 77.2 (#16/50)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 83% | 80.7% | GPT-5.6 Sol |
| MMMU-Pro w/ Python | 84.6% | 82% | GPT-5.6 Sol |

### Knowledge

- Winner: GPT-5.6 Sol
- GPT-5.6 Sol public-lane score: 78.8 (Supported · #7/158)
- GPT-5.6 Terra public-lane score: 71.1 (Supported · #9/158)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 94.6% | 92.9% | GPT-5.6 Sol |
| GPQA-D | 94.6% | 92.9% | GPT-5.6 Sol |
| HLE-Verified | 54.5% | 51.1% | GPT-5.6 Sol |
| LABBench2 | 82.1% | 81.2% | GPT-5.6 Sol |
| HealthBench Professional | 60.5% | 57.7% | GPT-5.6 Sol |
| HealthBench Hard | 33.1% | 32.7% | GPT-5.6 Sol |
| GPQA Diamond (Vals) | 95.2% | 90.9% | GPT-5.6 Sol |
| MMLU-Pro (Vals) | 89.1% | 86.7% | GPT-5.6 Sol |

### Instruction Following

- Winner: Directional only
- GPT-5.6 Sol public-lane score: 87.7 (#28/124)
- GPT-5.6 Terra public-lane score: 85.7 (#35/124)

### Mathematics

- Winner: Not comparable
- GPT-5.6 Sol public-lane score: 96.8 (Unranked · 3 rankable rows)
- GPT-5.6 Terra public-lane score: 96.8 (Unranked · 3 rankable rows)

| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath (legacy) | 89% | 84.9% | GPT-5.6 Sol |
| FrontierMath v2 (Tiers 1-3) | 89.000% | 84.900% | GPT-5.6 Sol |
| FrontierMath v2 (Tier 4) | 83.000% | 68.300% | GPT-5.6 Sol |

## FAQ

### Which is better overall, GPT-5.6 Sol or GPT-5.6 Terra?

GPT-5.6 Sol is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-5.6 Sol and GPT-5.6 Terra?

The widest category gap is in multimodal & grounded, where the averages are 87.6 for GPT-5.6 Sol and 77.2 for GPT-5.6 Terra.

### How many benchmarks does GPT-5.6 Sol cover on BenchLM?

GPT-5.6 Sol currently has 37 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-5.6 Terra cover on BenchLM?

GPT-5.6 Terra currently has 32 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.6 Sol vs GPT-5.6 Luna](/compare/gpt-5-6-luna-vs-gpt-5-6-sol)
- [GPT-5.6 Terra vs GPT-5.6 Luna](/compare/gpt-5-6-luna-vs-gpt-5-6-terra)
- [GPT-5.6 Sol vs GPT-5.6 Cyber](/compare/gpt-5-6-cyber-vs-gpt-5-6-sol)
- [GPT-5.6 Terra vs GPT-5.6 Cyber](/compare/gpt-5-6-cyber-vs-gpt-5-6-terra)

## Explore More

- [GPT-5.6 Sol profile](/models/gpt-5-6-sol)
- [GPT-5.6 Terra profile](/models/gpt-5-6-terra)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)