k
keynews.ai
keynews.ai
FrontierThe story, in brief

Sakana claims its AI model router Fugu Ultra v1.1 now beats Fable 5 without even including it in the pool

7.9 points. That's how much Sakana's Fugu Ultra v1.1 claims to improve — and it doesn't even need Fable 5 in the pool to beat it.

Reporting by The Decoder 1 min brief
Illustration of a transparent lens revealing connected networks across layers of paper.
Exploring the next frontier of AI research.AI illustration by KeyNews
The KeyNews take

Why it matters

Sakana's model router upgrade claims significant performance gains, but lacks independent verification. The Claude Code integration signals growing ecosystem consolidation around agentic tooling, though the unverified benchmark claim and EU unavailability limit immediate business impact.

The key facts

12 to know

  1. Fugu Ultra v1.1 claims 7.9-point improvement over v1.0

  2. Claims to beat Fable 5 without including it in routing pool

  3. Independent verification does not yet exist

  4. Adds Claude Code-compatible endpoint

  5. Service remains unavailable in EU

  6. UNVERIFIED — benchmark claims lack third-party validation

  7. Fugu Ultra v1.1 claims up to 7.9-point improvement over v1.0

  8. Claims to outperform Fable 5 without including it in the routing pool

  9. No independent verification of benchmark claims yet

  10. Added Claude Code-compatible endpoint

  11. Service unavailable in EU

  12. UNVERIFIED — claims lack third-party validation

Go to the source

The Decoderthe-decoder.com

Publisher excerpt: Sakana AI has updated its Fugu Ultra AI router to version 1.1, claiming gains of up to 7.9 points over v1.0. Independent verification doesn't exist yet. The update adds a Claude Code-compatible endpoint. The service remains unavailable in the EU.
Read original report
Back to today's editionMore frontier news

The wider picture

View all
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier01

Alibaba Unveils Zhenwu V900 — and Plans Qwen Models With Up to 10 Trillion Parameters

Alibaba is advancing on two fronts simultaneously: announcing a custom AI accelerator (Zhenwu V900) and committing to massive model scale (10T parameters for future Qwen releases). For practitioners, this matters as a credible third-party capability play outside the US-China licensing squeeze; for enthusiasts, it's a significant lab-race signal about training compute and parameter scaling as competitive levers.

TechRepublic
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier02

Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning

A new capability frontier: models that reason directly in speech without transcription bottlenecks. This changes how we think about multimodal reasoning and what's possible with open-weight releases at scale.

MarkTechPost
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Frontier03

Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war

Frontier labs are shipping upgraded reasoning and multimodal models in rapid succession, signaling acceleration in the capability race. Simultaneous price cuts reshape AI economics for practitioners.

Simon Willison