# GPT-5.6 Luna vs Inkling-Small

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

- Canonical page: https://benchlm.ai/compare/gpt-5-6-luna-vs-inkling-small
- Last updated: September 10, 2026

- Shared sourced benchmarks: 12
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

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

## Summary

- GPT-5.6 Luna leads overall 64.65 to 59.25.
- The clearest category separation is in coding, where the averages are 66.8 for GPT-5.6 Luna and 43.8 for Inkling-Small.
- The biggest single benchmark swing is Terminal-Bench 2.1 (Vals) in Agentic, with scores of 79.0% and 55.1%.
- GPT-5.6 Luna is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.6 Luna also has the larger context window at 1.05M.

## Model Snapshot

| Property | GPT-5.6 Luna | Inkling-Small |
|----------|----------|----------|
| Creator | OpenAI | Thinking Machines Lab |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Hybrid |
| Context | 1.05M | 1M |
| Overall Score | 64.65 | 59.25 |
| Benchmarks Covered | 30 | 25 |
| Pricing (input/output) | $0.20 / $1.20 | $0.58 / $1.44 |

## Category Breakdown

### Agentic

- Winner: GPT-5.6 Luna
- GPT-5.6 Luna public-lane score: 56.8 (Supported · #32/152)
- Inkling-Small public-lane score: 36.9 (Supported · #128/152)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 3.0 | 14.3% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 84.7% | 64.7% | GPT-5.6 Luna |
| BrowseComp | 83.3% | 77.4% | GPT-5.6 Luna |
| OSWorld 2.0 | 45.6% | Coming soon | Coming soon |
| CyberGym | 77.9% | Coming soon | Coming soon |
| ExploitGym | 12.4% | Coming soon | Coming soon |
| Toolathlon | 53.4% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 79.0% | 55.1% | GPT-5.6 Luna |
| ApprenticeBench | 7% | Coming soon | Coming soon |
| MCP Atlas | Coming soon | 79.6% | Coming soon |
| Toolathlon-Verified | Coming soon | 54.4% | Coming soon |

### Coding

- Winner: GPT-5.6 Luna
- GPT-5.6 Luna public-lane score: 66.8 (Supported · #9/151)
- Inkling-Small public-lane score: 43.8 (Supported · #97/151)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Pro | 62.7% | 55.9% | GPT-5.6 Luna |
| Terminal-Bench 2.0 | 84.7% | 64.7% | GPT-5.6 Luna |
| DeepSWE | 67.2% | Coming soon | Coming soon |
| FrontierCode 1.1 Extended | 55.1% | Coming soon | Coming soon |
| cursorBench32 | 61.1% | Coming soon | Coming soon |
| VulcanBench v3 | 85.5% | Coming soon | Coming soon |
| SWE-bench (Vals) | 93.0% | 82.2% | GPT-5.6 Luna |
| cursorBench40 | 35.9% | Coming soon | Coming soon |
| SWE-bench Verified | Coming soon | 80.2% | Coming soon |
| SciCode | Coming soon | 48.7% | Coming soon |
| LiveCodeBench (Vals) | Coming soon | 85.9% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- GPT-5.6 Luna public-lane score: 67.1 (#21/48)
- Inkling-Small public-lane score: 48.8 (#39/48)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 78.4% | 74% | GPT-5.6 Luna |
| MMMU-Pro w/ Python | 79.5% | Coming soon | Coming soon |
| CharXiv | Coming soon | 81.3% | Coming soon |
| CharXiv w/o tools | Coming soon | 77.4% | Coming soon |

### Reasoning

- Winner: Coming soon
- GPT-5.6 Luna public-lane score: 51.2 (#19/20)
- Inkling-Small public-lane score: 42.7 (Unranked · 3 rankable rows)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 59.5% | 40.1% | GPT-5.6 Luna |
| ARC-AGI-3 | 0.2% | Coming soon | Coming soon |
| CritPt | Coming soon | 8.3% | Coming soon |

### Knowledge

- Winner: GPT-5.6 Luna
- GPT-5.6 Luna public-lane score: 64.1 (Supported · #27/183)
- Inkling-Small public-lane score: 56.5 (Supported · #46/183)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.3% | 89.5% | GPT-5.6 Luna |
| GPQA-D | 92.3% | 89.5% | GPT-5.6 Luna |
| HealthBench Professional | 55.7% | Coming soon | Coming soon |
| HealthBench Hard | 32.0% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 91.7% | 83.6% | GPT-5.6 Luna |
| MMLU-Pro (Vals) | 86.0% | 85.6% | GPT-5.6 Luna |
| HLE | Coming soon | 47.8% | Coming soon |
| HLE w/o tools | Coming soon | 31.6% | Coming soon |

### Instruction Following

- Winner: Coming soon
- GPT-5.6 Luna public-lane score: Coming soon
- Inkling-Small public-lane score: 89.6 (#25/123)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| IFBench | Coming soon | 82.2% | Coming soon |

### Mathematics

- Winner: Coming soon
- GPT-5.6 Luna public-lane score: 94.4 (Unranked · 3 rankable rows)
- Inkling-Small public-lane score: 76.9 (Unranked · 2 rankable rows)

| Benchmark | GPT-5.6 Luna | Inkling-Small | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath (legacy) | 78.6% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 78.600% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 58.500% | Coming soon | Coming soon |
| AIME26 | Coming soon | 95.5% | Coming soon |
| HMMT Feb 2026 | Coming soon | 90.2% | Coming soon |

## FAQ

### Which is better overall, GPT-5.6 Luna or Inkling-Small?

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

### Where is the biggest gap between GPT-5.6 Luna and Inkling-Small?

The widest category gap is in coding, where the averages are 66.8 for GPT-5.6 Luna and 43.8 for Inkling-Small.

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

GPT-5.6 Luna currently has 30 sourced benchmark scores on BenchLM.

### How many benchmarks does Inkling-Small cover on BenchLM?

Inkling-Small currently has 25 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.6 Luna vs GPT-5.6 Sol](/compare/gpt-5-6-luna-vs-gpt-5-6-sol)
- [Inkling-Small vs GPT-5.6 Sol](/compare/gpt-5-6-sol-vs-inkling-small)
- [GPT-5.6 Luna vs GPT-5.6 Terra](/compare/gpt-5-6-luna-vs-gpt-5-6-terra)
- [Inkling-Small vs GPT-5.6 Terra](/compare/gpt-5-6-terra-vs-inkling-small)

## Explore More

- [GPT-5.6 Luna profile](/models/gpt-5-6-luna)
- [Inkling-Small profile](/models/inkling-small)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)