Last verified
RETIRING MAY 25 2026131K CONTEXTTEXT ONLYPROMPT CACHINGK2 ARCHIVE

Kimi K2 (0711 Preview) API Pricing

Kimi K2 (0711 Preview) is a deprecated original July K2 preview with the smaller 128K-class context window. The live Kimi pricing surface lists $0.6/M input and $2.5/M output, with cache hits at $0.15/M; Kimi says the whole kimi-k2 series retires on May 25, 2026. Pulled directly from platform.kimi.ai daily.

Input - per 1M tokens
$0.60/M
Source Kimi API retiring
Output - per 1M tokens
$2.50/M
standard K2 preview retiring
Cached input - per 1M tokens
$0.15/M
Cache automatic -75%
Effective - agentic blend
$0.41/M
92/8 split - 82% cache
§ 01 / TERMINAL

Run the numbers.

Live calculator pre-loaded with Kimi K2 (0711 Preview) archive rates. Use it for invoice checks and migration sizing, then compare against Kimi K2.5 before new production work.

$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS

Real-world presets.

CODING AGENT

Legacy codebase patch

$0.027/task
28,000 in - 4,000 out~3,731 units/$100
DEEP RESEARCH

Tool-heavy brief

$0.053/brief
55,000 in - 8,000 out~1,886 units/$100
RAG

Knowledge-base lookup

$0.008/query
9,000 in - 1,200 out~11,904 units/$100
CHATBOT

Support turn

$0.004/turn
3,500 in - 700 out~25,641 units/$100
§ 03 / TAPE

Price history.

Kimi K2 (0711 Preview) remains at $0.6/M input and $2.5/M output before the May 25 retirement.

Input · $0.60/M
Output · $2.5/M
Cached · $0.15/M
JUL 11 Launch pricing at $0.6/M input and $2.5/M outputMAY 22 Verified as deprecated; Kimi retires the kimi-k2 series on 2026-05-25
§ 04 / TOKENIZER

Paste text. See tokens. See cost.

Estimate · moonshot-tokenizer-estimate · ≈3.85 chars/token Auto-counts as you type

This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.

Characters —
Words —
Tokens (estimated) —
Cost as input · uncached —
Cost as output · uncached —
Cost as cached input —
§ 05 / SHELF

Up against the shelf.

All models →
Model Input /M Output /M Effective blended Context Best for
Kimi K2 (0711 Preview) Current $0.60 cache $0.15 $2.50 $0.41 agentic 92/8 131K Deprecated K2 archive page
Kimi K2.5 $0.60 cache $0.10 $3.00 $0.41 same blend 262K Supported K2 migration target
Kimi K2.6 $0.95 cache $0.16 $4.00 $0.60 pricier 262K Current Kimi flagship agents
Gemini 2.5 Flash $0.30 cache $0.03 $2.50 $0.27 cheaper 1M Multimodal budget work
Doubao Seed 2.0 Pro $0.45 cache $0.09 $2.25 $0.32 cheaper 256K Chinese multimodal agents
GPT-5.4 mini $0.75 cache $0.07 $4.50 $0.54 pricier 400K OpenAI low-latency coding

Frequently asked.

Archive pricing questions for teams still seeing this slug in logs, invoices, or migration plans.

Q · 01 Is Kimi K2 (0711 Preview) still supported? +
No for new planning. Kimi's pricing page says the kimi-k2 series will be discontinued on May 25, 2026 and recommends moving to newer Kimi models. This page keeps the price visible for archive and invoice checks.
Q · 02 What is Kimi K2 (0711 Preview) priced at? +
Kimi K2 (0711 Preview) is listed at $0.6/M input, $2.5/M output, and $0.15/M cache-hit input. Prices exclude applicable taxes and are stored here in USD per one million tokens.
Q · 03 Which model should replace it? +
Use Kimi K2.5 for new Kimi workloads. It is the recommended migration target for this archive slug and remains on the current Kimi pricing surface.
Q · 04 How does prompt caching affect the cost? +
Kimi supports automatic context caching for K2 models. Cache-hit input is billed at $0.15/M instead of the fresh-input $0.6/M rate, and AI//COST's default effective blend assumes an 82% cache-hit rate.
Q · 05 How is the effective price calculated? +
AI//COST uses the same 92/8 agentic blend everywhere. With an 82% cache-hit rate, Kimi K2 (0711 Preview)'s effective blended cost is $0.41/M.
Q · 06 Does this K2 variant support vision? +
No. Kimi documents the K2 preview series as text-only and explicitly says it does not support vision functionality. Use Kimi K2.5 or Kimi K2.6 when image or video input is part of the workload.
Q · 07 Are taxes or regional surcharges included? +
No. The Kimi pricing page states that listed prices exclude applicable taxes and checkout applies tax based on jurisdiction. AI//COST stores the vendor token price before taxes.
Y
Reviewed by Yaroslav Vikhariev Founder - AI//COST - Pricing pulled daily from platform.kimi.ai - Last verified May 22, 2026
Methodology Report a correction More by Y.V.