Model comparison
Sakana Fugu vs Sakana Fugu-Ultra
Head-to-head evidence from 11 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Sibling matchup inside the Sakana Fugu family.
Evidence parity. Sakana Fugu and Sakana Fugu-Ultra share 11 comparable benchmark results. 5 of 8 categories are comparable. 0 results are unique to Sakana Fugu; 0 to Sakana Fugu-Ultra.
Updated July 23, 2026- Shared results
- 11
- Sakana Fugu only
- 0
- Sakana Fugu-Ultra only
- 0
- Comparable categories
- 5 / 8
Sakana Fugu makes more sense if you want this variant’s specific profile, while Sakana Fugu-Ultra is the cleaner fit if reasoning is the priority.
Confidence note. This is a partial-evidence comparison with 11 shared benchmark results across 5 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Sakana Fugu and Sakana Fugu-Ultra sit in the same Sakana Fugu family. This page is less about two unrelated model lineages and more about how the siblings trade off on benchmark shape, token costs, and practical limits like context window.
Sakana Fugu and Sakana Fugu-Ultra finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Sakana Fugu | Δ | Sakana Fugu-Ultra |
|---|---|---|---|
| Reasoning | Sakana Fugu86.6 | Margin→ 7.0 | Sakana Fugu-Ultra93.6 |
| Coding | Sakana Fugu59.7 | Margin→ 4.8 | Sakana Fugu-Ultra64.5 |
| Agentic | Sakana Fugu80.2 | Margin→ 1.9 | Sakana Fugu-Ultra82.1 |
| Multimodal | Sakana Fugu85.1 | Margin→ 1.5 | Sakana Fugu-Ultra86.6 |
| Knowledge | Sakana Fugu95.5 | MarginTie | Sakana Fugu-Ultra95.5 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
SWE-bench Pro
CodingA 59%B 73.7%Winner: Sakana Fugu-UltraΔ 14.7SWE-bench Pro: Sakana Fugu scored 59%; Sakana Fugu-Ultra scored 73.7%. Sakana Fugu-Ultra wins this benchmark. - Source ↗
MRCRv2
ReasoningA 86.6%B 93.6%Winner: Sakana Fugu-UltraΔ 7MRCRv2: Sakana Fugu scored 86.6%; Sakana Fugu-Ultra scored 93.6%. Sakana Fugu-Ultra wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 80.2%B 82.1%Winner: Sakana Fugu-UltraΔ 1.9Terminal-Bench 2.0: Sakana Fugu scored 80.2%; Sakana Fugu-Ultra scored 82.1%. Sakana Fugu-Ultra wins this benchmark. - Source ↗
CharXiv
MultimodalA 85.1%B 86.6%Winner: Sakana Fugu-UltraΔ 1.5CharXiv: Sakana Fugu scored 85.1%; Sakana Fugu-Ultra scored 86.6%. Sakana Fugu-Ultra wins this benchmark. - Source ↗
SciCode
CodingA 60.1%B 58.7%Winner: Sakana FuguΔ 1.4SciCode: Sakana Fugu scored 60.1%; Sakana Fugu-Ultra scored 58.7%. Sakana Fugu wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Sakana Fugu | Sakana Fugu-Ultra | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Sakana FuguNot available | Sakana Fugu-UltraNot available | A complete price comparison is not available. |
| Generation speedtokens per second | Sakana FuguNot available | Sakana Fugu-UltraNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Sakana FuguNot available | Sakana Fugu-UltraNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Sakana Fugu1M | Sakana Fugu-Ultra1M | Listed context windows are equal. |
Benchmark Deep Dive
AgenticSakana Fugu-Ultra wins1 benchmarks
| Benchmark | Sakana Fugu | Sakana Fugu-Ultra | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 80.2% | 82.1% | Sakana Fugu-Ultra leads |
CodingSakana Fugu-Ultra wins5 benchmarks
| Benchmark | Sakana Fugu | Sakana Fugu-Ultra | Result |
|---|---|---|---|
| SWE-bench ProSource | 59% | 73.7% | Sakana Fugu-Ultra leads |
| Terminal-Bench 2.0Source | 80.2% | 82.1% | Sakana Fugu-Ultra leads |
| LiveCodeBench v6Source | 92.9% | 93.2% | Sakana Fugu-Ultra leads |
| LiveCodeBench ProSource | 87.8% | 90.8% | Sakana Fugu-Ultra leads |
| SciCodeSource | 60.1% | 58.7% | Sakana Fugu leads |
ReasoningSakana Fugu-Ultra wins1 benchmarks
| Benchmark | Sakana Fugu | Sakana Fugu-Ultra | Result |
|---|---|---|---|
| MRCRv2Source | 86.6% | 93.6% | Sakana Fugu-Ultra leads |
KnowledgeTie3 benchmarks
MultimodalSakana Fugu-Ultra wins1 benchmarks
| Benchmark | Sakana Fugu | Sakana Fugu-Ultra | Result |
|---|---|---|---|
| CharXivSource | 85.1% | 86.6% | Sakana Fugu-Ultra leads |
Frequently Asked Questions (6)
Which is better, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu and Sakana Fugu-Ultra are sibling variants in the Sakana Fugu family, so the right pick depends on whether you value the better benchmark line, cheaper tokens, or the larger context window. They are tied on the BenchAlign leaderboard on the current data.
Which is better for knowledge tasks, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu and Sakana Fugu-Ultra are effectively tied for knowledge tasks here, both landing at 95.5 on average.
Which is better for coding, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu-Ultra has the edge for coding in this comparison, averaging 64.5 versus 59.7. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.
Which is better for reasoning, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu-Ultra has the edge for reasoning in this comparison, averaging 93.6 versus 86.6. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu-Ultra has the edge for agentic tasks in this comparison, averaging 82.1 versus 80.2. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Sakana Fugu or Sakana Fugu-Ultra?
Sakana Fugu-Ultra has the edge for multimodal and grounded tasks in this comparison, averaging 86.6 versus 85.1. Inside this category, CharXiv is the benchmark that creates the most daylight between them.
Related Comparisons
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.