CADLE CAP
93
Cadle #21/385 · 52% confidence
sakana · ProprietarySakana: Fugu Ultra
Active
Model portfolio
Sakana: Fugu Ultra
Tap a star to rate
sakana · Released 2026-06-24 · v2026-06-24 · Closed · Context 1M tokens
Sakana: Fugu Ultra is a closed model from sakana (released 2026-06-24). Context window ~1M tokens with capabilities: Vision, Text. List pricing ≈ $5.00 / $30.00 per 1M input/output tokens. Top tracked signals: GPQA · benchlm 95.5, GPQA 95.5, REASON · benchlm 93.6. Cadle merges provider listings, unified leaderboard medians, and public pricing where available.
Language Cap
93
Cadle Language #21/385 · 52%
Benchmark average
87.5
#14 on the benchmark matrix
Pricing
$5.00
$30.00 / 1M out · 2.7 pts/$
CROSS-BOARD RANKS
Cadle Language LLM #21/385Matrix #14/443
CADLE CAP vs MODEL MATRIX
Cadle Cap ranks models on every published score we have for them — harder benchmarks weigh more. Missing tests are skipped, not zeroed. It answers who is stronger than whom in today's pool.
Model Matrix is the datasheet: Overall, Intelligence, Coding, Agentic, MMLU, GPQA, Math, LiveCodeBench, HumanEval, Reason, ARC-AGI. A dash means unmeasured. AVG is an importance-weighted mean of that model's filled cells only.
MODEL MATRIX · TRUTH CENTER CELLS
Overall
77.0
AA IQ
—
Coding
78.6
Agentic
82.1
MMLU
78.6
GPQA
95.5
Math
—
LiveCodeBench
93.2
HumanEval
—
Reason
93.6
ARC-AGI
—
Coverage AVG
87.5
Cadle Cap ranks models with within-benchmark percentiles so ARC-AGI ~40 and MMLU ~90 never compete raw. Coverage AVG is an importance-weighted mean of filled expert cells only — blanks are skipped, not zeroed.
ANALYSIS · TRUTH CENTER
Hard benches (GPQA · ARC-AGI · LCB · Agentic) sit left — a mid-40 on ARC-AGI can still be frontier. Broader benches on the right are easier and often higher.
CADLE INSIGHTS
Only measured signals
INTELLIGENCE
77.0
Media Cap quality axis
CODING
78.6
Media Cap quality axis
AGENTIC
82.1
Media Cap quality axis
KNOWLEDGE
89.7
Media Cap quality axis
MATRIX OVERALL
77.0
AVG 87.5 across matrix cells
BENCHMARK AVG
85.5
14 tracked scores
STRONGEST SIGNAL
GPQA · benchlm
Score 95.5
SPEED · COST · QUALITY
Measured matrix / AA / gateway rows only.
- Overall (Matrix)
- 77.0
- Cap Intelligence
- 95.0
- Cap Coding
- 89.3
- Matrix Coding
- 78.6
- Input $/1M
- $5.00
- Output $/1M
- $30.00
- Value pts/$
- 2.71
DESCRIPTION + DETAILS
Sakana: Fugu Ultra is a closed model from sakana (released 2026-06-24). Context window ~1M tokens with capabilities: Vision, Text. List pricing ≈ $5.00 / $30.00 per 1M input/output tokens. Top tracked signals: GPQA · benchlm 95.5, GPQA 95.5, REASON · benchlm 93.6. Cadle merges provider listings, unified leaderboard medians, and public pricing where available.
PROVIDER
sakana
RELEASE
2026-06-24
CONTEXT
1M tokens
STATUS
ACTIVE
COMMUNITY REVIEWS
OPERATOR NOTES
Quality, latency, cost, and failure modes — not marketing copy.
No reviews yet
WRITE A REVIEW
Tap a star to rate
Stored on this device · synced when signed in
No operator notes yet
First review sets the tone for how teams evaluate Sakana: Fugu Ultra.
MODEL COMPARISON
| Metric | |
|---|---|
| Provider | sakana |
| Release | 2026-06-24 |
| Input / 1M | $5.00 |
| Output / 1M | $30.00 |
| Context | 1M tokens |
| Cadle Cap | 93 |
| Cap board | Cadle Language #21/385 |
| GPQA · benchlm | 95.5 |
| GPQA | 95.5 |
| REASON · benchlm | 93.6 |
| REASON | 93.6 |
| LIVECODEBENCH · benchlm | 93.2 |
| LIVECODEBENCH | 93.2 |
| AGENTIC · benchlm | 82.1 |
| AGENTIC | 82.1 |
| CODING · benchlm | 78.6 |
| MMLU · benchlm | 78.6 |
| CODING | 78.6 |
| MMLU | 78.6 |
DATA PROVENANCE
Scores may use different scales — check the Benchmarks tab for source links.
- OpenRouter / provider APIs (pricing & context)
- Hugging Face Hub (downloads & cards)
- Papers With Code (task benchmarks)
- BenchLM.ai (category rollups + atomic benches)
- Epoch AI Benchmarking Hub (ECI + ZIP CSVs)
- LiveCodeBench (generation / execution / repair / test-output)
- Artificial Analysis (intelligence / indexes)
- LLM Stats (leaderboard + rankings)
- Cadle unified median (equal-weight multi-source)
- Epoch AI Benchmarking Hub — capability CSV exports (when synced)
SOURCE TRACKING
Evidence linked to this model — listing, hub, benchmarks, or live gateway prices.
API
sakana listing
API
OpenRouter
COMMUNITY
Papers with Code
COMMUNITY
BenchLM
COMMUNITY
unified-median
MEASURED CELLS
Same Truth Center numbers as the datasheet above. Accuracy, reliability, safety, and cost scores are not invented.
Overall77
Coding78.6
Agentic82.1
MMLU78.6
GPQA95.5
LiveCodeBench93.2