Sakana Fugu-Ultra
Sakana AI · Proprietary · bench-align-v5
Capability shape
Seven axes from the ranking source. A missing axis is drawn as a gap.
Runtime service evidence
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: ObservedUnavailable
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | Unavailable | Unavailable | Unavailable | |
| Throughput | Unavailable | Unavailable | Unavailable |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
Endpoint and price matrix
Every published price component, including cache reads and writes.
Workload-aware monthly cost example
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Benchmark record
11 matched benchmark rows with their published value, unit, and provenance.
Knowledge
| Benchmark | Value | Tasks | Difficulty | Provenance |
|---|---|---|---|---|
| GPQA Graduate-Level Google-Proof Q&A | 95.5 | 448 questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| GPQA-D GPQA Diamond | 95.5 | Graduate-level science questions | Graduate level | Trinity-Large-Thinking: Scaling an Open Source Frontier Agent |
| HLE w/o tools Humanity's Last Exam without tools | 50 | Expert-level questions | Frontier expert level | Introducing GPT-5.4 mini and nano |
Coding
| Benchmark | Value | Tasks | Difficulty | Provenance |
|---|---|---|---|---|
| Terminal-Bench 2.0 Terminal-Bench 2.0 | 82.1 | Terminal-based software tasks | Professional software engineering | Terminal-Bench 2.0 |
| LiveCodeBench v6 LiveCodeBench v6 | 93.2 | Fresh programming problems | Competitive programming level | LiveCodeBench official repository and release documentation |
| LiveCodeBench Pro LiveCodeBench Pro | 90.8 | Quarter-specific contest programming sets | High-end contest programming | LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? |
| SWE-bench Pro SWE-bench Pro | 73.7 | 1,865 repository problems | Long-horizon professional engineering | SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks? |
| SciCode Scientific Code Benchmark | 58.7 | — | — | BenchLM |
Reasoning
| Benchmark | Value | Tasks | Difficulty | Provenance |
|---|---|---|---|---|
| MRCRv2 MRCRv2 | 93.6 | Long-context retrieval | Hard long-context | Introducing GPT-5.2 and GPT-5.2 Pro |
Agentic
| Benchmark | Value | Tasks | Difficulty | Provenance |
|---|---|---|---|---|
| Terminal-Bench 2.0 Terminal-Bench 2.0 | 82.1 | Terminal-based software tasks | Professional software engineering | Terminal-Bench 2.0 |
Multimodal & Grounded
| Benchmark | Value | Tasks | Difficulty | Provenance |
|---|---|---|---|---|
| CharXiv CharXiv Reasoning | 86.6 | Scientific chart reasoning | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
Lifecycle and limitations log
Lifecycle events the source associates with this model, plus what this profile does not claim.
That is not evidence the model has no lifecycle plan — only that this source published none.