> Put up to four AI models side by side on the same independent tests, with API price and context window. Every comparison has its own link.

Up to four models

# Compare

Side by side on the same independent tests, with price and context. The link in your address bar shares this exact comparison.

[Claude Opus 5.5](https://themodelindex.org/models/claude-opus-5-5/)

[Kimi K3](https://themodelindex.org/models/kimi-k3/)

Add model

Next to Claude Opus 5.5

Closest rival, another lab

GPT-6 Astra

The version it replaced

Claude Opus 5

Where it started

GPT-4

Presets

Best open vs best closed (measured)

Today vs GPT-4 (2023)

Top three labs

US vs China, best of each

[Claude Opus 5.5](https://themodelindex.org/models/claude-opus-5-5/)

200

188–211

Weights

Closed

Released

Sep 2026

Price in / out

$4.0 / $20

Context

1M

Core tests taken

4 of 8

[Kimi K3](https://themodelindex.org/models/kimi-k3/)

172

161–183

Weights

Open

Released

Jul 2026

Price in / out

$0.68 / $10

Context

1.05M

Core tests taken

4 of 8

Claude Opus 5.5 leads Kimi K3 by 27 points, ahead on 3 of 4 shared core tests, at 2.7× the price.

## Test by test · independent results · Epoch AI

Claude Opus 5.5

Kimi K3

_Chart: Test results for the selected models_

Only tests that at least two of these models have taken. Bold rows are the core tests shown on every profile; all of these results inform the index score. Task length (METR) is listed below the chart because it is a duration, not a percentage.

---
Source: https://themodelindex.org/compare/ · The Model Index · data as of 2026-10-01