sakana/fugu-max

Model release OpenRouter 模型 Source language: Chinese

OpenRouter 于 2026-09-11 收录该模型(收录时间不等于模型正式发布日期);上下文:1,000,000 tokens;输入价:$2.0000/百万 tokens;输出价:$6.0000/百万 tokens;定位(来源描述):Fugu Max is the cost-performance model in Sakana AI's Fugu family.

Full text

Sakana: Fugu Max

sakana/fugu-max

Compare

Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route tasks across a fixed pool of open-weights and specialized models, including NVIDIA's Nemotron family, and to recursively call instances of itself.

Fugu Max dynamically selects efficient combinations of expert agents to improve quality and cost together, and is priced flat regardless of context length. It supports configurable reasoning effort (high, xhigh, max), function calling, structured outputs, image and PDF input, and built-in web search and web fetch. Orchestration tokens consumed by the system are billed as standard input/output tokens.

Modalities

In / out price

$2 / $6per 1M

Context

1.0M

Released

Sep 11, 2026

Compare

Providers

This model is hosted by one provider. OpenRouter forwards every request without routing.

Latency / throughput

$2.00$6.00$0.252.65s75 tok/s

100.00%

ProviderInput / Output /M

Input $2.00 / Output $6.00

75 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$1.522/M tokens

Output

$6/M tokens

Fugu Max — Price history

Price history

No effective price data for this selection

$1.52$6.0027.3%100.0%

ProviderInput / Output /M

Token share

Effective in $1.52 / Effective out $6.00

Token share

100.0%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

75tok/s

P50, best across providers

Latency

2.65s

P50, best provider

Reasoning effort

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

100.00%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

91.45%

Availability over the last 3 days

Last 72 hours

Availability 91.45%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

82.40%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Zed Editor

AI code editor designed for high-performance collaboration

2.29Btokens

2.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

243Mtokens

3.

Nous Research API

Nous Research is an AI research team focused on advancing open-source AI.

82.4Mtokens

4.

Vessel

new

65.9Mtokens

5.

Chatbox AI

On-device AI copilot

31.6Mtokens

Rest of OpenRouter

49K

Top 5 Apps

0

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Not currently available through EU in-region routing. In-region routing docs

Frequently asked questions

Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route tasks across a fixed pool of open-weights and specialized models, including NVIDIA's Nemotron family, and to recursively call instances of itself.

Fugu Max costs $2.00/M input tokens and $6.00/M output tokens, with separate rates for Cache Read at $0.25/M tokens and Web Search at $10.00/1K calls.

Fugu Max has a 1,000,000 token context window. It supports up to 128,000 completion tokens.

Yes. Fugu Max accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Fugu Max accepts text, images, and files such as PDFs as input and returns text.

Fugu Max is served by 1 provider on OpenRouter: Sakana.

Fugu Ultra v2, Sakana Namazu, and Fugu Ultra are other text models from Sakana.

Fugu Max was released on September 11, 2026.

Sources

← Back to the daily brief