Skip to main content
BenchLM
Advertise

Model comparison

Grok 4.5 vs Sakana Fugu-Ultra

Data verified

Head-to-head evidence from 3 shared benchmark results across 2 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

76.72/100
No comparison
N/A
2 category wins0 category wins

Public leaderboard positions: Grok 4.5 #7 (Supported); Sakana Fugu-Ultra unranked (Not scored). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Grok 4.5 and Sakana Fugu-Ultra share 3 comparable benchmark results. 2 of 8 categories are comparable. 25 results are unique to Grok 4.5; 8 to Sakana Fugu-Ultra.

Updated July 23, 2026
Shared results
3
Grok 4.5 only
25
Sakana Fugu-Ultra only
8
Comparable categories
2 / 8

Treat this as a split decision. Grok 4.5 makes more sense if agentic is the priority; Sakana Fugu-Ultra is the better fit if you need the larger 1M context window.

Confidence note. This is a partial-evidence comparison with 3 shared benchmark results across 2 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Grok 4.5 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 500K for Grok 4.5.

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 Grok 4.5 and Sakana Fugu-Ultra
CategoryGrok 4.5ΔSakana Fugu-Ultra
AgenticGrok 4.583.3Margin← 1.2Sakana Fugu-Ultra82.1
CodingGrok 4.564.7Margin← 0.2Sakana Fugu-Ultra64.5
ReasoningGrok 4.5Not measuredMarginNo overlapSakana Fugu-Ultra93.6
KnowledgeGrok 4.5Not measuredMarginNo overlapSakana Fugu-Ultra95.5
MultimodalGrok 4.5Not measuredMarginNo overlapSakana Fugu-Ultra86.6

Decisive benchmark drivers

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

More
A · Grok 4.5B · Sakana Fugu-Ultra

  1. SWE-bench Pro

    Coding
    Source ↗
    A 64.7%B 73.7%
    Winner: Sakana Fugu-UltraΔ 9
    SWE-bench Pro: Grok 4.5 scored 64.7%; Sakana Fugu-Ultra scored 73.7%. Sakana Fugu-Ultra wins this benchmark.

  2. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 83.3%B 82.1%
    Winner: Grok 4.5Δ 1.2
    Terminal-Bench 2.0: Grok 4.5 scored 83.3%; Sakana Fugu-Ultra scored 82.1%. Grok 4.5 wins this benchmark.

Operational comparison

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

MetricGrok 4.5Sakana Fugu-UltraComparison
Input / output priceUSD per 1M tokensGrok 4.5$2 input / $6 outputSakana Fugu-UltraNot availableA complete price comparison is not available.
Generation speedtokens per secondGrok 4.5Not availableSakana Fugu-UltraNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGrok 4.5Not availableSakana Fugu-UltraNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGrok 4.5500KSakana Fugu-Ultra1MSakana Fugu-Ultra lists the larger context window.

Benchmark Deep Dive

AgenticGrok 4.5 wins
BenchmarkGrok 4.5Sakana Fugu-UltraResult
Terminal-Bench 2.0Source 83.3%82.1%Grok 4.5 leads
deepSweSource 53%—Not comparable
AA Agentic IndexSource 45.7%—Not comparable
GDPval-AASource 51.7%—Not comparable
GDPval-AASource 1535—Not comparable
AA BriefcaseSource 1323—Not comparable
AA AutomationBenchSource 51.4%—Not comparable
AA EnterpriseOps-GymSource 40.8%—Not comparable
AA Harvey LABSource 92.4%—Not comparable
AA Tau3 BankingSource 32.6%—Not comparable
aaTerminalBench21Source 81.6%—Not comparable
CodingGrok 4.5 wins
BenchmarkGrok 4.5Sakana Fugu-UltraResult
SWE-bench ProSource 64.7%73.7%Sakana Fugu-Ultra leads
SWE MultilingualSource 78%—Not comparable
Terminal-Bench 2.0Source 83.3%82.1%Grok 4.5 leads
cursorBench32Source 66.7%—Not comparable
VulcanBench v3Source 91.3%—Not comparable
AA Coding IndexSource 72.5%—Not comparable
AA-SciCodeSource 54.1%—Not comparable
LiveCodeBench v6Source —93.2%Not comparable
LiveCodeBench ProSource —90.8%Not comparable
SciCodeSource —58.7%Not comparable
Reasoning
BenchmarkGrok 4.5Sakana Fugu-UltraResult
AA-LCRSource 67.7%—Not comparable
CritPtSource 15.4%—Not comparable
MRCRv2Source —93.6%Not comparable
Knowledge
BenchmarkGrok 4.5Sakana Fugu-UltraResult
Artificial Analysis Intelligence IndexSource 53.8%—Not comparable
AA-GPQA DiamondSource 93.1%—Not comparable
AA-HLESource 40.3%—Not comparable
AA-Omniscience IndexSource 26.4%—Not comparable
AA-Omniscience AccuracySource 52.1%—Not comparable
AA-Omniscience Hallucination RateSource 53.5%—Not comparable
GPQASource —95.5%Not comparable
GPQA-DSource —95.5%Not comparable
HLE w/o toolsSource —50%Not comparable
Multimodal
BenchmarkGrok 4.5Sakana Fugu-UltraResult
AA-MMMU-ProSource 80.4%—Not comparable
Design Arena WebsiteSource 1325—Not comparable
CharXivSource —86.6%Not comparable
Frequently Asked Questions (3)

Which is better, Grok 4.5 or Sakana Fugu-Ultra?

Grok 4.5 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 coding, Grok 4.5 or Sakana Fugu-Ultra?

Grok 4.5 has the edge for coding in this comparison, averaging 64.7 versus 64.5. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Grok 4.5 or Sakana Fugu-Ultra?

Grok 4.5 has the edge for agentic tasks in this comparison, averaging 83.3 versus 82.1. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

xAI Compare Pricing Methodology Find Your Best LLM Overall Rankings
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.