AI Index

Claude Opus 5.5vsKimi K3

Both language models, compared at each maker’s own API list prices. Every figure links to its source.

Compare at your usage

Every model priced for the same request
Prompt per request
Cache hits
Batch

8,000 tokens in, 2,000 out

Claude Opus 5.5 on Claude APIKimi K3 on Moonshot AI Platform
Rate appliedBase rateBase rate
Inputper 1M tokens$4.00 $3.00
Cache readper 1M tokens$0.20 $0.30
Outputper 1M tokens$20.00 $15.00
Cost per request$0.072$0.054
Per 1,000 requests$72.00$54.00

Every model is priced for the same request at its maker’s own list prices (each with its source and date), so bold marks the cheapest. A model that can’t serve the request is named, not ranked; a condition its seller doesn’t price says which rate was used.

Cost per request as prompts grow

2K tokens out, list price on each maker’s API.

0$1.00$2.00$3.00$4.000250K500K750K1Mprompt tokens per request8KClaude Opus 5.5$0.072 at 8KKimi K3$0.054 at 8K
  • Claude Opus 5.5 · Claude API
  • Kimi K3 · Moonshot AI Platform

Where each stands among all scored models

Each line runs from the lowest to the highest score on the board; the tick is the median.

Epoch Capabilities Index#3 vs #21 of 118Kimi K3: 158.5Claude Opus 5.5: 168.4LMArena Text#3 vs #15 of 163Kimi K3: 1,488Claude Opus 5.5: 1,509LMArena Text · Coding#3 vs #11 of 163Kimi K3: 1,540Claude Opus 5.5: 1,551LMArena Text · Chinese#3 vs #10 of 161Kimi K3: 1,543Claude Opus 5.5: 1,574
  • Claude Opus 5.5
  • Kimi K3

Key differences

Claude Opus 5.5Kimi K3
InputMaker’s official API · global · USD
$4.00 / 1M tokens Claude API pricing · as of Nov 1, 2026
$3.00 / 1M tokens Moonshot AI platform pricing · as of Nov 1, 2026
OutputMaker’s official API · global · USD
$20.00 / 1M tokens Claude API pricing · as of Nov 1, 2026
$15.00 / 1M tokens Moonshot AI platform pricing · as of Nov 1, 2026
Context
1,000,000 Claude Opus 5.5 model overview · as of Nov 1, 2026
1,048,576 Moonshot AI platform pricing · as of Nov 1, 2026
Max output
128,000 Claude Opus 5.5 model overview · as of Nov 1, 2026
1,048,576 Kimi API chat completions reference · as of Nov 1, 2026
Input modalities
Text, Image Claude Opus 5.5 model overview · as of Nov 1, 2026
Text, Image, Video moonshotai/Kimi-K3 model card · as of Oct 30, 2026
Cache readMaker’s official API · global · USD
$0.20 / 1M tokens Claude API pricing · as of Nov 1, 2026
$0.30 / 1M tokens Moonshot AI platform pricing · as of Nov 1, 2026

These differences use verified vendor specifications or each maker’s official API on the same region, currency, meter and billing conditions. Different conditions remain side by side in the full table; they do not establish which model is cheaper. Declared features describe support, not measured quality or a recommendation. Missing evidence does not mean a feature is unsupported.

Full comparison

Claude Opus 5.5 Anthropic Kimi K3 Moonshot AI
Overview
ReleasedSep 22, 2026 Jul 2026
Open weights— Yes
Knowledge cutoffJun 2026 Not stated
Specs
Context1M 1.05M
Max output128K 1.05M
Input modalitiesText, Image Text, Image, Video
Output modalitiesText Text
ReasoningYes Yes
Reasoning effortsmax low, high, max
Reasoning can be offNot stated Yes
Tool useYes Yes
Structured outputYes Yes
Pricing · per 1M tokens
Input price$4.00Also priced under other conditions: Batch $3.00
Output price$20.00Also priced under other conditions: Batch $15.00
Cache read price$0.20 $0.30
Cache write price$5.00Also priced under other conditions: 1-hour cache5-minute cache $3.00Also priced under other conditions: 1-hour cache5-minute cache
Cache write price (1-hour cache)$8.001-hour cache $6.001-hour cache
Blended price (3 input : 1 output)$8.00 $6.00
Pricing under conditions · per 1M tokens
Batch input price$2.00Batch Not offered
Batch output price$10.00Batch Not offered
Cited scores
Epoch Capabilities Index168.4 158.5
LMArena Text1,509 high effort1,488 max effort
LMArena Text · Coding1,551 high effort1,540 max effort
LMArena Vision1,321 high effort—
LMArena Text · Chinese1,574 high effort1,543 max effort
Epoch AI · GPQA Diamond90.6% max effort93.1% max effort
Epoch AI · FrontierMath Tiers 1–391.2% max effort72.2% max effort

Headline prices may come from different channels, regions or tiers. Check each source before choosing a model; missing facts stay blank.

Model details and sources

Cited scores