# GPT-5.4 mini vs Kimi K2.5 (Reasoning)

> Side-by-side benchmark comparison for GPT-5.4 mini and Kimi K2.5 (Reasoning) across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-4-mini-vs-kimi-k2-5-reasoning
- Last updated: September 10, 2026

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

## Quick Verdict

Pick GPT-5.4 mini if you want the stronger benchmark profile. Kimi K2.5 (Reasoning) only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.4 mini leads overall 61.11 to 57.
- The clearest category separation is in agentic, where the averages are 39.1 for GPT-5.4 mini and 48.2 for Kimi K2.5 (Reasoning).
- The biggest single benchmark swing is Vibe Code Bench in Coding, with scores of 47.97% and 17.54%.
- Kimi K2.5 (Reasoning) is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.4 mini also has the larger context window at 400K.

## Model Snapshot

| Property | GPT-5.4 mini | Kimi K2.5 (Reasoning) |
|----------|----------|----------|
| Creator | OpenAI | Moonshot AI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 400K | 128K |
| Overall Score | 61.11 | 57 |
| Benchmarks Covered | 19 | 9 |
| Pricing (input/output) | $0.75 / $4.50 | $0.60 / $3.00 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 39.1 (Estimated · #119/152)
- Kimi K2.5 (Reasoning) public-lane score: 48.2 (Estimated · #65/152)

| Benchmark | GPT-5.4 mini | Kimi K2.5 (Reasoning) | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 60% | 50.8% | GPT-5.4 mini |
| OSWorld-Verified | 72.1% | Coming soon | Coming soon |
| MCP Atlas | 57.7% | Coming soon | Coming soon |
| Toolathlon | 42.9% | Coming soon | Coming soon |
| τ²-bench results | 93.4% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 54.7% | Coming soon | Coming soon |
| BrowseComp | Coming soon | 60.6% | Coming soon |
| Gert Labs | Coming soon | 32.58% | Coming soon |

### Coding

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 42.7 (Supported · #104/151)
- Kimi K2.5 (Reasoning) public-lane score: 50.7 (Estimated · #53/151)

| Benchmark | GPT-5.4 mini | Kimi K2.5 (Reasoning) | Winner |
|-----------|-----------|-----------|--------|
| Vibe Code Bench | 47.97% | 17.54% | GPT-5.4 mini |
| FrontierCode 1.1 Main | 27.0% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 81.5% | Coming soon | Coming soon |
| SWE-bench (Vals) | 73.0% | Coming soon | Coming soon |
| SWE-bench Verified | Coming soon | 76.8% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 57.2 (#31/48)
- Kimi K2.5 (Reasoning) public-lane score: 63.2 (Unranked · 1 rankable row)

| Benchmark | GPT-5.4 mini | Kimi K2.5 (Reasoning) | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 76.6% | 78.5% | Kimi K2.5 (Reasoning) |
| MMMU-Pro w/ Python | 78% | Coming soon | Coming soon |

### Reasoning

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 73.9 (Unranked · 2 rankable rows)
- Kimi K2.5 (Reasoning) public-lane score: Coming soon

### Knowledge

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 55.6 (Supported · #52/183)
- Kimi K2.5 (Reasoning) public-lane score: 50.7 (Estimated · #75/183)

| Benchmark | GPT-5.4 mini | Kimi K2.5 (Reasoning) | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 88% | 87.6% | GPT-5.4 mini |
| HLE | 41.5% | Coming soon | Coming soon |
| HLE w/o tools | 28.2% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 83.1% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 84.6% | Coming soon | Coming soon |
| MMLU-Pro | Coming soon | 87.1% | Coming soon |

### Instruction Following

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 89.8 (#23/123)
- Kimi K2.5 (Reasoning) public-lane score: Coming soon

### Mathematics

- Winner: Coming soon
- GPT-5.4 mini public-lane score: 44.5 (Unranked · 2 rankable rows)
- Kimi K2.5 (Reasoning) public-lane score: Coming soon

| Benchmark | GPT-5.4 mini | Kimi K2.5 (Reasoning) | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 28.280% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 2.080% | Coming soon | Coming soon |
| AIME 2025 | Coming soon | 96.1% | Coming soon |

## FAQ

### Which is better overall, GPT-5.4 mini or Kimi K2.5 (Reasoning)?

GPT-5.4 mini is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-5.4 mini and Kimi K2.5 (Reasoning)?

The widest category gap is in agentic, where the averages are 39.1 for GPT-5.4 mini and 48.2 for Kimi K2.5 (Reasoning).

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

GPT-5.4 mini currently has 19 sourced benchmark scores on BenchLM.

### How many benchmarks does Kimi K2.5 (Reasoning) cover on BenchLM?

Kimi K2.5 (Reasoning) currently has 9 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.4 mini vs GPT-5.4](/compare/gpt-5-4-vs-gpt-5-4-mini)
- [Kimi K2.5 (Reasoning) vs GPT-5.4](/compare/gpt-5-4-vs-kimi-k2-5-reasoning)
- [GPT-5.4 mini vs GPT-5.4 Pro](/compare/gpt-5-4-mini-vs-gpt-5-4-pro)
- [Kimi K2.5 (Reasoning) vs GPT-5.4 Pro](/compare/gpt-5-4-pro-vs-kimi-k2-5-reasoning)

## Explore More

- [GPT-5.4 mini profile](/models/gpt-5-4-mini)
- [Kimi K2.5 (Reasoning) profile](/models/kimi-k2-5-reasoning)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)