Model comparison
Sakana Fugu-Ultra vs Trinity-Large-Thinking
Head-to-head evidence from 1 shared benchmark result across 1 category. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Sakana Fugu-Ultra unranked (Not scored); Trinity-Large-Thinking #100 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Sakana Fugu-Ultra and Trinity-Large-Thinking share 1 comparable benchmark result. 0 of 8 categories are comparable. 10 results are unique to Sakana Fugu-Ultra; 18 to Trinity-Large-Thinking.
Updated July 23, 2026- Shared results
- 1
- Sakana Fugu-Ultra only
- 10
- Trinity-Large-Thinking only
- 18
- Comparable categories
- 0 / 8
Benchmark data for Sakana Fugu-Ultra and Trinity-Large-Thinking is coming soon on BenchLM.
Confidence note. This is a partial-evidence comparison with 1 shared benchmark result across 1 evidence category; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.
Sakana Fugu-Ultra has the larger context window at 1M, compared with 512K for Trinity-Large-Thinking.
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-Ultra | Δ | Trinity-Large-Thinking |
|---|---|---|---|
| Agentic | Sakana Fugu-Ultra82.1 | MarginNo overlap | Trinity-Large-ThinkingNot measured |
| Coding | Sakana Fugu-Ultra64.5 | MarginNo overlap | Trinity-Large-ThinkingNot measured |
| Reasoning | Sakana Fugu-Ultra93.6 | MarginNo overlap | Trinity-Large-ThinkingNot measured |
| Knowledge | Sakana Fugu-Ultra95.5 | MarginNo overlap | Trinity-Large-ThinkingNot measured |
| Multimodal | Sakana Fugu-Ultra86.6 | MarginNo overlap | Trinity-Large-ThinkingNot measured |
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Sakana Fugu-Ultra | Trinity-Large-Thinking | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Sakana Fugu-UltraNot available | Trinity-Large-Thinking$0.25 input / $0.9 output | A complete price comparison is not available. |
| Generation speedtokens per second | Sakana Fugu-UltraNot available | Trinity-Large-ThinkingNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Sakana Fugu-UltraNot available | Trinity-Large-ThinkingNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Sakana Fugu-Ultra1M | Trinity-Large-Thinking512K | Sakana Fugu-Ultra lists the larger context window. |
Benchmark Deep Dive
Agentic5 benchmarks
Coding7 benchmarks
| Benchmark | Sakana Fugu-Ultra | Trinity-Large-Thinking | Result |
|---|---|---|---|
| SWE-bench ProSource | 73.7% | — | Not comparable |
| Terminal-Bench 2.0Source | 82.1% | — | Not comparable |
| LiveCodeBench v6Source | 93.2% | — | Not comparable |
| LiveCodeBench ProSource | 90.8% | — | Not comparable |
| SciCodeSource | 58.7% | — | Not comparable |
| SWE-bench Verified*Source | — | 63.2% | Not comparable |
| AA-SciCodeSource | — | 36.1% | Not comparable |
Reasoning3 benchmarks
Knowledge10 benchmarks
| Benchmark | Sakana Fugu-Ultra | Trinity-Large-Thinking | Result |
|---|---|---|---|
| GPQASource | 95.5% | — | Not comparable |
| GPQA-DSource | 95.5% | 76.3% | Sakana Fugu-Ultra leads |
| HLE w/o toolsSource | 50% | — | Not comparable |
| MMLU-Pro (Arcee)Source | — | 83.4% | Not comparable |
| Artificial Analysis Intelligence IndexSource | — | 24.5% | Not comparable |
| AA-GPQA DiamondSource | — | 75.2% | Not comparable |
| AA-HLESource | — | 14.7% | Not comparable |
| AA-Omniscience IndexSource | — | -44.2% | Not comparable |
| AA-Omniscience AccuracySource | — | 22.8% | Not comparable |
| AA-Omniscience Hallucination RateSource | — | 86.6% | Not comparable |
Math1 benchmarks
| Benchmark | Sakana Fugu-Ultra | Trinity-Large-Thinking | Result |
|---|---|---|---|
| AIME25 (Arcee)Source | — | 96.3% | Not comparable |
Multimodal2 benchmarks
Inst. Following1 benchmarks
| Benchmark | Sakana Fugu-Ultra | Trinity-Large-Thinking | Result |
|---|---|---|---|
| AA-IFBenchSource | — | 56.3% | Not comparable |
Frequently Asked Questions (3)
Can I compare Sakana Fugu-Ultra and Trinity-Large-Thinking on BenchLM yet?
Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.
Why does this comparison show “coming soon”?
BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.
What data is available for Sakana Fugu-Ultra and Trinity-Large-Thinking today?
Sakana Fugu-Ultra: Pricing unavailable Trinity-Large-Thinking: $0.25 input / $0.90 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.
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.