# GPT-5.4 nano vs Sakana Fugu-Ultra

> Side-by-side benchmark comparison for GPT-5.4 nano and Sakana Fugu-Ultra across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-4-nano-vs-sakana-fugu-ultra
- Last updated: September 28, 2026

- Shared sourced benchmarks: 2
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Evidence is too thin for a benchmark verdict: GPT-5.4 nano and Sakana Fugu-Ultra share only 2 sourced benchmark rows.

## Summary

- Only 2 shared sourced benchmark rows are available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Sakana Fugu-Ultra has the lower published output-token price.
- Sakana Fugu-Ultra has the larger published context window at 1M.

## Model Snapshot

| Property | GPT-5.4 nano | Sakana Fugu-Ultra |
|----------|----------|----------|
| Creator | OpenAI | Sakana AI |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 400K | 1M |
| Independent public score (not head-to-head) | 50.8 | null |
| Benchmarks Covered | 20 | 11 |
| Pricing (input/output) | $0.20 / $1.25 | Pricing unavailable / Pricing unavailable |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 31.8 (Supported · #75/117)
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 46.3% | Coming soon | Coming soon |
| OSWorld-Verified | 39% | Coming soon | Coming soon |
| MCP Atlas | 56.1% | Coming soon | Coming soon |
| Toolathlon | 35.5% | Coming soon | Coming soon |
| τ²-bench results | 92.5% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 41.6% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 82.1% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 31.2 (Supported · #90/142)
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| Vibe Code Bench | 26.10% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 84.0% | Coming soon | Coming soon |
| SWE-bench (Vals) | 69.8% | Coming soon | Coming soon |
| SWE-bench Pro | Coming soon | 73.7% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 82.1% | Coming soon |
| LiveCodeBench v6 | Coming soon | 93.2% | Coming soon |
| LiveCodeBench Pro | Coming soon | 90.8% | Coming soon |
| SciCode | Coming soon | 58.7% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 37 (Unranked · 4 rankable rows)
- Sakana Fugu-Ultra public-lane score: 79.1 (Unranked · 1 rankable row)

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-1 | 51.50% | Coming soon | Coming soon |
| ARC-AGI-2 | 5.7% | Coming soon | Coming soon |
| MRCRv2 | Coming soon | 93.6% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 24.9 (#47/50)
- Sakana Fugu-Ultra public-lane score: 73.6 (Unranked · 1 rankable row)

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 66.1% | Coming soon | Coming soon |
| MMMU-Pro w/ Python | 69.5% | Coming soon | Coming soon |
| CharXiv | Coming soon | 86.6% | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 37.8 (Supported · #104/168)
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 82.8% | 95.5% | Sakana Fugu-Ultra |
| HLE | 37.7% | Coming soon | Coming soon |
| HLE w/o tools | 24.3% | 50% | Sakana Fugu-Ultra |
| GPQA Diamond (Vals) | 77.5% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 77.2% | Coming soon | Coming soon |
| GPQA-D | Coming soon | 95.5% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 91.9 (#9/124)
- Sakana Fugu-Ultra public-lane score: Coming soon

### Mathematics

- Winner: Insufficient shared evidence
- GPT-5.4 nano public-lane score: 43.8 (Unranked · 2 rankable rows)
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | GPT-5.4 nano | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 25.860% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 6.250% | Coming soon | Coming soon |

## FAQ

### Is there enough evidence to pick between GPT-5.4 nano and Sakana Fugu-Ultra?

Not yet. The models share only 2 sourced benchmark rows, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [GPT-5.4 nano vs GPT-5.4 Pro](/compare/gpt-5-4-nano-vs-gpt-5-4-pro)
- [Sakana Fugu-Ultra vs GPT-5.4 Pro](/compare/gpt-5-4-pro-vs-sakana-fugu-ultra)
- [GPT-5.4 nano vs GPT-5.4](/compare/gpt-5-4-vs-gpt-5-4-nano)
- [Sakana Fugu-Ultra vs GPT-5.4](/compare/gpt-5-4-vs-sakana-fugu-ultra)

## Explore More

- [GPT-5.4 nano profile](/models/gpt-5-4-nano)
- [Sakana Fugu-Ultra profile](/models/sakana-fugu-ultra)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)