# GPT-5.4 vs Kimi K3.1 Pro

> Side-by-side benchmark comparison for GPT-5.4 and Kimi K3.1 Pro across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-4-vs-kimi-3-1-pro
- Last updated: October 23, 2026

- Shared sourced benchmarks: 11
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Pick Kimi K3.1 Pro if you want the stronger benchmark profile. GPT-5.4 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Kimi K3.1 Pro leads overall 82 to 66.56.
- The clearest category separation is in multimodal & grounded, where the averages are 71.5 for GPT-5.4 and 98 for Kimi K3.1 Pro.
- The biggest single benchmark swing is HLE in Knowledge, with scores of 52.1% and 70%.
- Kimi K3.1 Pro is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | GPT-5.4 | Kimi K3.1 Pro |
|----------|----------|----------|
| Creator | OpenAI | Moonshot AI |
| Type | Proprietary | Pending |
| Reasoning | Reasoning | Reasoning |
| Context | 1.05M | 1.05M |
| Overall Score | 66.56 | null |
| Benchmarks Covered | 43 | 42 |
| Pricing (input/output) | $2.50 / $15.00 | $2.00 / $10.00 |

## Category Breakdown

### Agentic

- Winner: Not comparable
- GPT-5.4 public-lane score: 52.4 (Supported · #46/132)
- Kimi K3.1 Pro public-lane score: Coming soon

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 75.1% | Coming soon | Coming soon |
| CyberGym | 79.0% | Coming soon | Coming soon |
| BrowseComp | 82.7% | 92.1% | Kimi K3.1 Pro |
| OSWorld-Verified | 75% | Coming soon | Coming soon |
| MCP Atlas | 70.6% | Coming soon | Coming soon |
| Toolathlon | 54.6% | Coming soon | Coming soon |
| τ²-bench results | 98.9% | Coming soon | Coming soon |
| Claw-Eval | 60.3% | Coming soon | Coming soon |
| DeepSearchQA | 73.6% | Coming soon | Coming soon |
| Gert Labs | 64.89% | Coming soon | Coming soon |
| ResearchClawBench | 15.3% | Coming soon | Coming soon |
| JobBench | 38.9% | Coming soon | Coming soon |
| ExploitGym | 6.0% | 20.7% | Kimi K3.1 Pro |
| ApprenticeBench | 11% | Coming soon | Coming soon |
| MazeBench | 0% | Coming soon | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 65.70% | Coming soon |
| Terminal-Bench-Science 0.1 | Coming soon | 38.1% | Coming soon |
| HLE w/ tools | Coming soon | 70.0% | Coming soon |
| BioMysteryBench | Coming soon | 80.6% | Coming soon |
| Agents' Last Exam | Coming soon | 43.2% | Coming soon |
| Agents' Last Exam (CLI) | Coming soon | 44.2% | Coming soon |
| APEX-Agents | Coming soon | 67.2% | Coming soon |
| Frontier Finance | Coming soon | 58.6% | Coming soon |
| HANDBOOK.md | Coming soon | 44.6% | Coming soon |
| OSWorld 2.0 | Coming soon | 79.6% | Coming soon |
| Business Operation | Coming soon | 37.8% | Coming soon |
| Financial Modeling | Coming soon | 57.6% | Coming soon |
| Financial Research | Coming soon | 70.4% | Coming soon |
| Kimi Hallu Bench Hard (tools) | Coming soon | 8.3% | Coming soon |

### Coding

- Winner: Not comparable
- GPT-5.4 public-lane score: 47.7 (Supported · #50/153)
- Kimi K3.1 Pro public-lane score: Coming soon

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench Pro | 87.5% | Coming soon | Coming soon |
| SWE-bench Pro | 57.7% | Coming soon | Coming soon |
| React Native Evals | 85.3% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 75.1% | Coming soon | Coming soon |
| Vibe Code Bench | 67.42% | Coming soon | Coming soon |
| PostTrainBench v1.1 | 19.0% | Coming soon | Coming soon |
| DeepSWE | Coming soon | 75.5% | Coming soon |
| ProgramBench (binary) | Coming soon | 31.5% | Coming soon |
| Kimi Code Bench v2 | Coming soon | 77.1% | Coming soon |
| Kimi Code Bench 3.0 Long-horizon | Coming soon | 61.2% | Coming soon |
| sweMarathon | Coming soon | 51.7% | Coming soon |
| PostTrainBench v1.2 | Coming soon | 36.5% | Coming soon |

### Reasoning

- Winner: Not comparable
- GPT-5.4 public-lane score: 64.2 (#20/27)
- Kimi K3.1 Pro public-lane score: Coming soon

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 74.0% | Coming soon | Coming soon |
| ARC-AGI-3 | 0.2% | Coming soon | Coming soon |
| ARC-AGI-1 | 93.67% | Coming soon | Coming soon |

### Multimodal & Grounded

- Winner: Kimi K3.1 Pro
- GPT-5.4 public-lane score: 71.5 (#33/56)
- Kimi K3.1 Pro public-lane score: 98 (#1/56)

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| ParseBench | 62.23% | Coming soon | Coming soon |
| MMMU-Pro | 81.2% | 84.1% | Kimi K3.1 Pro |
| OfficeQA Pro | 53.2% | 67.4% | Kimi K3.1 Pro |
| MMMU-Pro w/ Python | 82.1% | 90% | Kimi K3.1 Pro |
| CharXiv | 82.8% | 93.4% | Kimi K3.1 Pro |
| ERQA | 65.4% | Coming soon | Coming soon |
| SimpleVQA | 61.1% | Coming soon | Coming soon |
| ScreenSpot Pro | 85.4% | Coming soon | Coming soon |
| ZeroBench | 41.0% | 39.0% | GPT-5.4 |
| MedXpertQA (MM) | 77.1% | Coming soon | Coming soon |
| OfficeQA Pro v2 | Coming soon | 64.4% | Coming soon |
| BenchCAD Vision2Code (no tools) | Coming soon | 0.728 | Coming soon |
| BenchCAD Vision2Code (tools) | Coming soon | 0.927 | Coming soon |
| CharXiv w/o tools | Coming soon | 89% | Coming soon |
| Chartography (tools) | Coming soon | 90.2% | Coming soon |
| BabyVision w/ Python | Coming soon | 95.4% | Coming soon |
| ZeroBench w/ Python | Coming soon | 63.0% | Coming soon |
| GDP.pdf (Gemini judge) | Coming soon | 34.6% | Coming soon |
| GDP.pdf w/ Python (Gemini judge) | Coming soon | 85.7% | Coming soon |
| PerceptionBench | Coming soon | 68.3% | Coming soon |

### Knowledge

- Winner: Not comparable
- GPT-5.4 public-lane score: 68.5 (Supported · #21/187)
- Kimi K3.1 Pro public-lane score: Coming soon

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.8% | 94.2% | Kimi K3.1 Pro |
| HLE | 52.1% | 70% | Kimi K3.1 Pro |
| HLE w/o tools | 39.8% | 51.1% | Kimi K3.1 Pro |
| GPQA-D | 92.8% | 94.2% | Kimi K3.1 Pro |
| HealthBench Hard | 40.1% | Coming soon | Coming soon |
| MedXpertQA (Text) | 59.6% | Coming soon | Coming soon |
| HealthBench Professional | 48.1% | Coming soon | Coming soon |
| Kimi Hallu Bench | Coming soon | 5.4% | Coming soon |

### Instruction Following

- Winner: Not comparable
- GPT-5.4 public-lane score: 90.7 (#25/153)
- Kimi K3.1 Pro public-lane score: Coming soon

### Mathematics

- Winner: Not comparable
- GPT-5.4 public-lane score: 82 (Unranked · 2 rankable rows)
- Kimi K3.1 Pro public-lane score: Coming soon

| Benchmark | GPT-5.4 | Kimi K3.1 Pro | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 78.596% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 49.000% | Coming soon | Coming soon |

## FAQ

### Which is better overall, GPT-5.4 or Kimi K3.1 Pro?

Kimi K3.1 Pro is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-5.4 and Kimi K3.1 Pro?

The widest category gap is in multimodal & grounded, where the averages are 71.5 for GPT-5.4 and 98 for Kimi K3.1 Pro.

### How many benchmarks does GPT-5.4 cover on BenchLM?

GPT-5.4 currently has 43 sourced benchmark scores on BenchLM.

### How many benchmarks does Kimi K3.1 Pro cover on BenchLM?

Kimi K3.1 Pro currently has 42 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.4 vs GPT-5.4 Pro](/compare/gpt-5-4-vs-gpt-5-4-pro)
- [Kimi K3.1 Pro vs GPT-5.4 Pro](/compare/gpt-5-4-pro-vs-kimi-3-1-pro)
- [GPT-5.4 vs GPT-5.4 mini](/compare/gpt-5-4-vs-gpt-5-4-mini)
- [Kimi K3.1 Pro vs GPT-5.4 mini](/compare/gpt-5-4-mini-vs-kimi-3-1-pro)

## Explore More

- [GPT-5.4 profile](/models/gpt-5-4)
- [Kimi K3.1 Pro profile](/models/kimi-3-1-pro)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)