Kimi K2 Thinking
by Moonshot AI · moonshotai/kimi-k2-thinking · #189 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:36 AM UTC · refreshed hourly
Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.60
per 1M tokens
Output
$2.50
per 1M tokens
Blended (3:1)
$1.07
3 input : 1 output
Cache read
$0.15
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Kimi K2 Thinking with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $1.07 | $0.0637 | — |
| 50% | $0.906 | $0.0412 | −35% |
| 90% | $0.771 | $0.0232 | −64% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.00105 | $1.05 |
| Document summary | 8,000 | 1,000 | $0.0073 | $7.30 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.023 | $23.00 |
| Long-context analysis | 150,000 | 5,000 | $0.1025 | $102.50 |
Performance
Intelligence Index
17.3
AA composite quality
Coding Index
21
Math Index
—
Agentic Index
1.8
Output speed
—
median
Time to first token
—
median
Time to first answer
—
after reasoning tokens
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,140 |
Specs
Context window
262K
tokens
Max output
100K
tokens
Input modalities
text
Tokenizer
Other