# GPT-5.2 vs Ornith-1.5-397B

> Side-by-side benchmark comparison for GPT-5.2 and Ornith-1.5-397B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-2-vs-ornith-1-5-397b
- Last updated: September 18, 2026

- Shared sourced benchmarks: 4
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Ornith-1.5-397B if you want the stronger benchmark profile. GPT-5.2 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Ornith-1.5-397B leads overall 65.32 to 64.86.
- The clearest category separation is in agentic, where the averages are 42.7 for GPT-5.2 and 60.9 for Ornith-1.5-397B.
- The biggest single benchmark swing is BrowseComp in Agentic, with scores of 65.8% and 86.6%.
- Ornith-1.5-397B is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.2 also has the larger context window at 400K.

## Model Snapshot

| Property | GPT-5.2 | Ornith-1.5-397B |
|----------|----------|----------|
| Creator | OpenAI | Ornith AI |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Reasoning |
| Context | 400K | 262K |
| Overall Score | 64.86 | 65.32 |
| Benchmarks Covered | 15 | 18 |
| Pricing (input/output) | $1.75 / $14.00 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- GPT-5.2 public-lane score: 42.7 (Supported · #107/154)
- Ornith-1.5-397B public-lane score: 60.9 (Estimated · #17/154)

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| BrowseComp | 65.8% | 86.6% | Ornith-1.5-397B |
| OSWorld-Verified | 47.3% | Coming soon | Coming soon |
| Gert Labs | 46.54% | Coming soon | Coming soon |
| JobBench | 34.3% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 86.1% | Coming soon |
| HLE w/ tools | Coming soon | 56.1% | Coming soon |
| MCP Atlas | Coming soon | 80% | Coming soon |
| Toolathlon-Verified | Coming soon | 71.2% | Coming soon |
| WideResearch | Coming soon | 80.8% | Coming soon |
| Claw-Eval | Coming soon | 81.4% | Coming soon |

### Coding

- Winner: Coming soon
- GPT-5.2 public-lane score: 46.3 (Supported · #87/154)
- Ornith-1.5-397B public-lane score: 59 (Estimated · #25/154)

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80% | 86% | Ornith-1.5-397B |
| SWE-bench Pro | 55.6% | 65.1% | Ornith-1.5-397B |
| Vibe Code Bench | 53.50% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 86.1% | Coming soon |
| SWE Multilingual | Coming soon | 79.6% | Coming soon |
| DeepSWE | Coming soon | 56.0% | Coming soon |
| frontierBench | Coming soon | 13.5% | Coming soon |
| NL2Repo | Coming soon | 59.5% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- GPT-5.2 public-lane score: 66.3 (#22/48)
- Ornith-1.5-397B public-lane score: Coming soon

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 79.5% | Coming soon | Coming soon |
| MathVision | 83.0% | Coming soon | Coming soon |
| CharXiv | 82.1% | Coming soon | Coming soon |
| V* | 75.9% | Coming soon | Coming soon |

### Reasoning

- Winner: Coming soon
- GPT-5.2 public-lane score: 53.7 (Unranked · 3 rankable rows)
- Ornith-1.5-397B public-lane score: Coming soon

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 52.9% | Coming soon | Coming soon |

### Knowledge

- Winner: Coming soon
- GPT-5.2 public-lane score: 61.6 (Supported · #29/184)
- Ornith-1.5-397B public-lane score: 60.5 (Estimated · #35/184)

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.4% | 92.8% | Ornith-1.5-397B |
| GPQA-D | Coming soon | 92.8% | Coming soon |
| HLE | Coming soon | 44.6% | Coming soon |
| HLE w/o tools | Coming soon | 44.6% | Coming soon |

### Instruction Following

- Winner: Coming soon
- GPT-5.2 public-lane score: 91.2 (#14/124)
- Ornith-1.5-397B public-lane score: Coming soon

### Mathematics

- Winner: Coming soon
- GPT-5.2 public-lane score: 57.4 (Unranked · 2 rankable rows)
- Ornith-1.5-397B public-lane score: Coming soon

| Benchmark | GPT-5.2 | Ornith-1.5-397B | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 40.700% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 18.800% | Coming soon | Coming soon |

## FAQ

### Which is better overall, GPT-5.2 or Ornith-1.5-397B?

Ornith-1.5-397B is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-5.2 and Ornith-1.5-397B?

The widest category gap is in agentic, where the averages are 42.7 for GPT-5.2 and 60.9 for Ornith-1.5-397B.

### How many benchmarks does GPT-5.2 cover on BenchLM?

GPT-5.2 currently has 15 sourced benchmark scores on BenchLM.

### How many benchmarks does Ornith-1.5-397B cover on BenchLM?

Ornith-1.5-397B currently has 18 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.2 vs GPT-5.2 Pro](/compare/gpt-5-2-vs-gpt-5-2-pro)
- [Ornith-1.5-397B vs GPT-5.2 Pro](/compare/gpt-5-2-pro-vs-ornith-1-5-397b)
- [GPT-5.2 vs GPT-5.2 Instant](/compare/gpt-5-2-vs-gpt-5-2-instant)
- [Ornith-1.5-397B vs GPT-5.2 Instant](/compare/gpt-5-2-instant-vs-ornith-1-5-397b)

## Explore More

- [GPT-5.2 profile](/models/gpt-5-2)
- [Ornith-1.5-397B profile](/models/ornith-1-5-397b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)