Skip to main content

Model comparison

MAI-Thinking-1 vs Sakana Fugu-Ultra

Data verified

Head-to-head evidence from 5 shared benchmark results across 3 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

N/A
No comparison
N/A
1 category wins2 category wins

Evidence parity. MAI-Thinking-1 and Sakana Fugu-Ultra share 5 comparable benchmark results. 3 of 8 categories are comparable. 8 results are unique to MAI-Thinking-1; 6 to Sakana Fugu-Ultra.

Updated July 23, 2026
Shared results
5
MAI-Thinking-1 only
8
Sakana Fugu-Ultra only
6
Comparable categories
3 / 8

Treat this as a split decision. MAI-Thinking-1 makes more sense if coding is the priority; Sakana Fugu-Ultra is the better fit if agentic is the priority or you need the larger 1M context window.

Confidence note. This is a partial-evidence comparison with 5 shared benchmark results across 3 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

MAI-Thinking-1 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.

Sakana Fugu-Ultra gives you the larger context window at 1M, compared with 256K for MAI-Thinking-1.

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 scores and score margins for MAI-Thinking-1 and Sakana Fugu-Ultra
CategoryMAI-Thinking-1ΔSakana Fugu-Ultra
AgenticMAI-Thinking-146.0Margin 36.1Sakana Fugu-Ultra82.1
KnowledgeMAI-Thinking-172.5Margin 23.0Sakana Fugu-Ultra95.5
CodingMAI-Thinking-165.5Margin 1.0Sakana Fugu-Ultra64.5
ReasoningMAI-Thinking-1Not measuredMarginNo overlapSakana Fugu-Ultra93.6
MathMAI-Thinking-189.7MarginNo overlapSakana Fugu-UltraNot measured
MultimodalMAI-Thinking-1Not measuredMarginNo overlapSakana Fugu-Ultra86.6
Inst. FollowingMAI-Thinking-185.0MarginNo overlapSakana Fugu-UltraNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · MAI-Thinking-1B · Sakana Fugu-Ultra
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 46%B 82.1%
    Winner: Sakana Fugu-UltraΔ 36.1
    Terminal-Bench 2.0: MAI-Thinking-1 scored 46%; Sakana Fugu-Ultra scored 82.1%. Sakana Fugu-Ultra wins this benchmark.
  2. SWE-bench Pro

    Coding
    Source ↗
    A 52.8%B 73.7%
    Winner: Sakana Fugu-UltraΔ 20.9
    SWE-bench Pro: MAI-Thinking-1 scored 52.8%; Sakana Fugu-Ultra scored 73.7%. Sakana Fugu-Ultra wins this benchmark.
  3. GPQA

    Knowledge
    Source ↗
    A 84.2%B 95.5%
    Winner: Sakana Fugu-UltraΔ 11.3
    GPQA: MAI-Thinking-1 scored 84.2%; Sakana Fugu-Ultra scored 95.5%. Sakana Fugu-Ultra wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricMAI-Thinking-1Sakana Fugu-UltraComparison
Input / output priceUSD per 1M tokensMAI-Thinking-1Not availableSakana Fugu-UltraNot availableA complete price comparison is not available.
Generation speedtokens per secondMAI-Thinking-1Not availableSakana Fugu-UltraNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenMAI-Thinking-1Not availableSakana Fugu-UltraNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensMAI-Thinking-1256KSakana Fugu-Ultra1MSakana Fugu-Ultra lists the larger context window.

Benchmark Deep Dive

AgenticSakana Fugu-Ultra wins
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
Terminal-Bench 2.0Source 46%82.1%Sakana Fugu-Ultra leads
CodingMAI-Thinking-1 wins
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
SWE-bench VerifiedSource 73.5%Not comparable
SWE-bench ProSource 52.8%73.7%Sakana Fugu-Ultra leads
Terminal-Bench 2.0Source 46.0%82.1%Sakana Fugu-Ultra leads
LiveCodeBench v6Source 93.2%Not comparable
LiveCodeBench ProSource 90.8%Not comparable
SciCodeSource 58.7%Not comparable
Reasoning
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
Graphwalks BFS 128KSource 90%Not comparable
MRCRv2Source 93.6%Not comparable
KnowledgeSakana Fugu-Ultra wins
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
GPQASource 84.2%95.5%Sakana Fugu-Ultra leads
GPQA-DSource 84.2%95.5%Sakana Fugu-Ultra leads
MMLU-ProSource 85%Not comparable
SimpleQASource 31%Not comparable
HLE w/o toolsSource 50%Not comparable
Math
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
AIME 2025Source 97%Not comparable
AIME26Source 94.5%Not comparable
HMMT Feb 2026Source 84.9%Not comparable
Multimodal
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
CharXivSource 86.6%Not comparable
Inst. Following
BenchmarkMAI-Thinking-1Sakana Fugu-UltraResult
IFBenchSource 85%Not comparable
Frequently Asked Questions (4)

Which is better, MAI-Thinking-1 or Sakana Fugu-Ultra?

MAI-Thinking-1 and Sakana Fugu-Ultra are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.

Which is better for knowledge tasks, MAI-Thinking-1 or Sakana Fugu-Ultra?

Sakana Fugu-Ultra has the edge for knowledge tasks in this comparison, averaging 95.5 versus 72.5. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, MAI-Thinking-1 or Sakana Fugu-Ultra?

MAI-Thinking-1 has the edge for coding in this comparison, averaging 65.5 versus 64.5. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, MAI-Thinking-1 or Sakana Fugu-Ultra?

Sakana Fugu-Ultra has the edge for agentic tasks in this comparison, averaging 82.1 versus 46. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 23, 2026

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.