LIVE

Your desk · free

Home Track record The AI map Regime Companies

Unlock with a seat

Ask Tessara House Views Constraint Radar Signals feed Catalyst calendar Earnings console Your Book · exposure Memory desk Compute desk

Unlock a seat →

Founding rate ends Jul 31

Tessara is independent research and analysis, not investment advice. We are not a registered investment adviser. Past performance is not indicative of future results, and you are responsible for your own investment decisions.

Terms Privacy Refund policy

← SignalsSignal

← SignalsSignal

Notable6.0BenchmarkModelsSingle sourceactive

Updated 5d ago · 1 source

Kimi-K3 Trails GPT-5.4-xhigh on LiveBench Benchmark

The Kimi-K3 model demonstrated performance behind GPT-5.4-xhigh on the LiveBench evaluation, indicating a competitive gap in model capabilities.

So What

A new benchmark result confirms GPT-5.4-xhigh's continued lead in model capabilities, setting a high bar for competitors.

Sources (1)

Kimi-K3 Trails GPT-5.4-xhigh on LiveBench Benchmark

General AI Twitter · Twitter · 5d ago

Related Signals

6d agoSignalKimi-K3 Benchmarks Reportedly Leaked, Signaling Potential Model Capabilities→6d agoBenchmarkKimi K3 Demonstrates Advanced Long-Duration AI Tasks, Challenges Claude and GPT→5d agoModel ReleaseKimi.ai Introduces Kimi K3 Open Frontier Intelligence Model→4d agoBenchmarkKimi K3 Ranks #5 on Artificial Analysis Coding Agent Index with Cost-Efficient Performance→6d agoSignalKimi K3 Features 2.8 Trillion Parameters, Exceeding GPT-5.6 and Opus 4.8→

Credibility

Lead sourceGeneral AI Twitter

Corroboration1 distinct outlet

VerificationSingle source

Unlock the full signal with Pro

Tessara synthesizes 80+ data sources into actionable intelligence for AI infrastructure investors.

Unlock with Pro→Sign in