# GPT-6.1 Terra vs GPT-6 Luna

> Side-by-side benchmark comparison for GPT-6.1 Terra and GPT-6 Luna across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-6-1-terra-vs-gpt-6-luna
- Last updated: October 26, 2026

- Shared sourced benchmarks: 6
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

GPT-6.1 Terra leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- GPT-6.1 Terra has the higher overall point estimate 75.87 to 66.4. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in reasoning, where the averages are 81.8 for GPT-6.1 Terra and 59.2 for GPT-6 Luna.
- The biggest single benchmark swing is OSWorld 2.1 Offline in Agentic, with scores of 74.2% and 48.9%.
- GPT-6 Luna is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | GPT-6.1 Terra | GPT-6 Luna |
|----------|----------|----------|
| Creator | OpenAI | OpenAI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 1.05M | 1.05M |
| Overall Score | 75.87 | 66.4 |
| Benchmarks Covered | 10 | 21 |
| Pricing (input/output) | $1.00 / $5.00 | $0.10 / $0.50 |

## Category Breakdown

### Agentic

- Winner: Directional only
- GPT-6.1 Terra public-lane score: 72.4 (Estimated · #8/133)
- GPT-6 Luna public-lane score: 58.3 (Estimated · #36/133)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 4.0 | 50.30% | Coming soon | Coming soon |
| Terminal-Bench-Science 0.1 | 58.3% | Coming soon | Coming soon |
| Agents' Last Exam | 60.0% | Coming soon | Coming soon |
| OSWorld 2.1 Offline | 74.2% | 48.9% | GPT-6.1 Terra |
| ExploitGym | Coming soon | 11.6% | Coming soon |
| Hermes Index | Coming soon | 33.9% | Coming soon |
| Hermes Bench | Coming soon | 70.5% | Coming soon |
| MazeBench | Coming soon | 0% | Coming soon |

### Coding

- Winner: Not comparable
- GPT-6.1 Terra public-lane score: Coming soon
- GPT-6 Luna public-lane score: 53.5 (Supported · #38/154)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| DeepSWE | Coming soon | 66.6% | Coming soon |
| Bug Hunt Bench | Coming soon | 18.3 fixes | Coming soon |

### Reasoning

- Winner: Not comparable
- GPT-6.1 Terra public-lane score: 81.8 (Unranked · 2 rankable rows)
- GPT-6 Luna public-lane score: 59.2 (#23/26)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-1 | Coming soon | 86.67% | Coming soon |
| ARC-AGI-2 | Coming soon | 59.3% | Coming soon |
| ARC-AGI-3 | Coming soon | 0.1% | Coming soon |
| ARC-AGI-3 Provider Adapter | Coming soon | 0.59% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- GPT-6.1 Terra public-lane score: 91.6 (#4/58)
- GPT-6 Luna public-lane score: 82.8 (Unranked · 1 rankable row)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 85.5% | Coming soon | Coming soon |
| ExtractBench Pro | Coming soon | 0% | Coming soon |
| ExtractBench (LlamaIndex) | Coming soon | 91.01% | Coming soon |
| ParseBench | Coming soon | 64.26% | Coming soon |

### Knowledge

- Winner: GPT-6.1 Terra
- GPT-6.1 Terra public-lane score: 74.9 (Supported · #12/188)
- GPT-6 Luna public-lane score: 66.7 (Supported · #28/188)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| HealthBench (raw) | 51.1% | 50.0% | GPT-6.1 Terra |
| HealthBench (length-adjusted) | 55.1% | 54.5% | GPT-6.1 Terra |
| HealthBench Professional | 62.7% | 60.8% | GPT-6.1 Terra |
| HealthBench Professional (raw) | 64.2% | 61.2% | GPT-6.1 Terra |
| HealthBench Hard | 31.4% | 31.4% | Tie |

### Mathematics

- Winner: Not comparable
- GPT-6.1 Terra public-lane score: Coming soon
- GPT-6 Luna public-lane score: 83.7 (Unranked · 2 rankable rows)

| Benchmark | GPT-6.1 Terra | GPT-6 Luna | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | Coming soon | 78.947% | Coming soon |
| FrontierMath v2 (Tier 4) | Coming soon | 56.098% | Coming soon |

## FAQ

### Which is better overall, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between GPT-6.1 Terra and GPT-6 Luna?

The widest category gap is in reasoning, where the averages are 81.8 for GPT-6.1 Terra and 59.2 for GPT-6 Luna.

### How many benchmarks does GPT-6.1 Terra cover on BenchLM?

GPT-6.1 Terra currently has 10 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-6 Luna cover on BenchLM?

GPT-6 Luna currently has 21 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-6.1 Terra vs GPT-6 Astra](/compare/gpt-6-1-terra-vs-gpt-6-astra)
- [GPT-6 Luna vs GPT-6 Astra](/compare/gpt-6-astra-vs-gpt-6-luna)
- [GPT-6.1 Terra vs GPT-6.1 Sol](/compare/gpt-6-1-sol-vs-gpt-6-1-terra)
- [GPT-6 Luna vs GPT-6.1 Sol](/compare/gpt-6-1-sol-vs-gpt-6-luna)

## Explore More

- [GPT-6.1 Terra profile](/models/gpt-6-1-terra)
- [GPT-6 Luna profile](/models/gpt-6-luna)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)