Skip to main content
AI Choice Engine

Sakana AI

Fugu Max v1.0

Fugu Max v1.0 is a Sakana AI model with Not verified context, $2 in / $6 out per million tokens, last verified October 2, 2026. Open weights: no. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Closed model-orchestration service that routes work among a pool of models to optimize cost and capability.

Recorded profile verification October 2, 2026Individual provider fields may change

AI model specification details
SpecificationFugu Max v1.0
ProviderSakana AI
TierBalanced
Context windowNot verifiedUnverified
Max outputNot verifiedUnverified
Input / 1M tokens$2
Output / 1M tokens$6
WeightsClosed
ParametersNot verifiedUnverified
Reasoning levelshigh, xhigh, max
Modalitiestext, image
API model idfugu-max-v1.0
ReleasedSeptember 11, 2026

Pricing tiers: Pinned fugu-max-v1.0 pay-as-you-go $2/$6 per MTok regardless of context length, cached input $0.25. Internal web_search/web_fetch cost $0.007 per call, and one query can invoke multiple calls. API-platform subscriptions are separate from Sakana Chat; subscription allowances are not per-token list prices.

Best for

  • Cost-performance routing
  • General agent work
  • Multi-model orchestration

Watch out

high/xhigh are distinct; max is accepted as an alias of xhigh. Endpoint defaults and ignored sampling controls are documented separately; send full history rather than previous_response_id. Publisher client metadata advertises 1M context and text/image input, but an exact server-enforced context/output maximum is unverified. EU/EEA access is unavailable. Provider headline benchmark claims are vendor results; no matched independent run or Ultra score is inherited.

Source receipts

Catalog figures for this model were checked against the following sources.

Common questions

Fugu Max v1.0

Answered from the verified figures on this page rather than general guidance.

How much does Fugu Max v1.0 cost per million tokens?
Fugu Max v1.0 lists base rates of $2 per million input tokens and $6 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. Pinned fugu-max-v1.0 pay-as-you-go $2/$6 per MTok regardless of context length, cached input $0.25. Internal web_search/web_fetch cost $0.007 per call, and one query can invoke multiple calls. API-platform subscriptions are separate from Sakana Chat; subscription allowances are not per-token list prices.
What is Fugu Max v1.0 best for?
Fugu Max v1.0 is a balanced tier from Sakana AI. It suits cost-performance routing, general agent work, multi-model orchestration. high/xhigh are distinct; max is accepted as an alias of xhigh. Endpoint defaults and ignored sampling controls are documented separately; send full history rather than previous_response_id. Publisher client metadata advertises 1M context and text/image input, but an exact server-enforced context/output maximum is unverified. EU/EEA access is unavailable. Provider headline benchmark claims are vendor results; no matched independent run or Ultra score is inherited.
Model comparison catalogBenchmark library