Breaking

Moonshot AI's Kimi K3 Tops Frontend Code Arena, Edging Out Anthropic's Claude Fable 5

July 16, 2026 at 15:28 EDT

Foundation Models
Software Dev & Coding
Open Source

Moonshot AI's newly released Kimi K3 has claimed the top spot on Arena.ai's Frontend Code Arena, narrowly surpassing Anthropic's Claude Fable 5 and marking a rare instance of an open-weight model leading a frontier coding leaderboard.

Jul 2026 · Moonshot AI

Kimi K3 Takes #1 on the Frontend Code Arena

Moonshot AI's open-weight Kimi K3 narrowly overtakes Anthropic's Claude Fable 5 — a rare instance of an open model leading a frontier coding leaderboard.

1,679

Kimi K3

open-weight · #1

1,631

Claude Fable 5

closed · #2

Frontend Code Arena score — Kimi K3 leads by 48 points

+17

places leapt from
Kimi-k2.6 (was #18)

2.8T

parameters, multimodal
reasoning model

token context
window

6/7

Arena domains
ranked #1

Domain-by-domain — Frontend Code Arena

✓ Brand & Marketing ✓ Reference-Based Design ✓ Data & Analytics ✓ Consumer Product ✓ Simulations ✓ Content Creation Tools #2 Gaming (behind Fable 5)

Where it shines

Visually ambitious frontend generation
3D simulations & detailed UIs
Visual edge over Fable 5 on some tasks
Complex coding, debugging, tool use

Where rivals still lead

Slower generation than Fable 5
UX component robustness can lag
Variable consistency, some task failures
Fable 5 & GPT-5.6 Sol keep speed edge

Access & pricing

Full open weights due July 27, 2026

API access opened ~July 16, 2026

per 1M input

$15

per 1M output

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.