Last verified
CLOSED TO NEW USERS OCT 31DEPRECATED256K CONTEXTTEXT + VISION + VIDEOPROMPT CACHINGOPENAI-COMPATIBLE API

Kimi K2.6 API Pricing

Deprecated: Moonshot closes the K2 series to newly registered users from October 31, 2026, keeps it for existing users, and plans to retire it by the end of December 2026; its migration guide points to Kimi K3.1 Pro. Kimi K2.6 is Moonshot's 2026 open-weight multimodal model for long-horizon coding, multimodal input and agent tasks. The live Kimi pricing surface lists $0.95/M input and $4/M output, with cache hits at $0.16/M. Checked by hand against platform.kimi.ai.

Input - per 1M tokens
$0.95/M
Source Kimi API flat
Output - per 1M tokens
$4.00/M
Context 256K flat
Cached input - per 1M tokens
$0.16/M
Cache automatic -83%
Effective - agentic blend
$0.598/M
92/8 split - 82% cache
§ 01 / TERMINAL

Kimi K2.6 API cost calculator.

Live calculator pre-loaded with Kimi K2.6 rates, which existing users pay until Moonshot retires the K2 series. Tweak spend, output mix, or cache assumptions and share the URL to share the calculation.

$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
Open full calculator (all models · share URL · CSV) →
§ 02 / SCENARIOS

Kimi K2.6 cost on real workloads.

CODING AGENT

Long-horizon coding

$0.030/task
32,000 in - 5,000 out~3,370 units/$100
VISION

Screenshot to implementation

$0.017/screen
18,000 in - 3,000 out~5,734 units/$100
DEEP RESEARCH

Research brief

$0.042/brief
45,000 in - 7,000 out~2,403 units/$100
CHATBOT

Support assistant

$0.00386/turn
3,500 in - 700 out~25,922 units/$100
§ 03 / TOKENIZER

Kimi K2.6 token counter and cost.

What's a token →
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 499
Words 72
Tokens (estimated) 130 tokens
Cost as input · uncached $0.00012 USD
Cost as output · uncached $0.00052 USD
Cost as cached input $0.00002 USD
§ 04 / SHELF

Kimi K2.6 vs similar models.

All models →
Model Input /M Output /M Effective blended Context Best for
Kimi K2.6 Current $0.95 cache $0.16 $4.00 $0.598 agentic 92/8 262K Frontier Kimi agents
Kimi K3 $3.00 cache $0.30 $15.00 $1.92 +3.2× 1M Multimodal Kimi production
Kimi K2 (0905 Preview) $0.60 cache $0.15 $2.50 $0.413 retired 262K Legacy K2 migration checks
Kimi K2 Turbo Preview $1.15 cache $0.15 $8.00 $0.944 retired 262K Legacy fast K2 traffic
GPT-6.1 Terra $1.00 cache $0.05 $5.00 $0.645 +8% 1.05M OpenAI's fast, efficient GPT-6 tier
Gemini 3.8 Flash $0.75 cache $0.075 $3.75 $0.481 −20% · promo 1M Mainstream multimodal work
Doubao Seed 2.0 Pro $0.476 cache $0.0952 $2.38 $0.341 deprecated · −43% 256K Chinese multimodal agents
DeepSeek V4.1 Flash $0.30 cache $0.006 $1.20 $0.15 −4.0× 1M Budget reasoning and coding
§ 06 / COMPARE

Compare Kimi K2.6.

All comparisons →
2-WAY
Kimi vs Claude
Anthropic
↗

Kimi K2.6 API pricing FAQ.

Practical pricing questions, separated from calculator assumptions and regional taxes.

Q · 01 Can I still use Kimi K2.6? +
If your account already exists, yes, at the rates on this page. Moonshot closes the K2 series to newly registered users from October 31, 2026, keeps it for existing users, and plans to retire it by the end of December 2026; its migration guide points to Kimi K3.1 Pro. Moonshot has not published an exact shutdown date. Batch still covers K2.6 at 60% of the standard price.
Q · 02 What is Kimi K2.6 priced at? +
Kimi K2.6 is listed at $0.95/M input, $4/M output, and $0.16/M cache-hit input. The Kimi pricing docs state that prices exclude applicable taxes.
Q · 03 How does prompt caching work? +
Kimi supports automatic context caching for this model. Cache-hit input is billed at $0.16/M instead of the fresh-input $0.95/M rate, and the calculator's default agentic blend assumes an 82% cache-hit rate.
Q · 04 How is the effective price calculated? +
AI//COST uses the same 92/8 agentic blend everywhere. With an 82% cache hit rate, Kimi K2.6's effective blended cost is $0.6/M.
Q · 05 Does this include image and video input? +
Kimi K2.6 is documented as a native multimodal model with text, image, and video input. The pricing page bills consumed tokens, so extracted document, image, or video context should be treated as model input once passed into a request.
Q · 06 Is there a batch discount? +
Yes. Kimi's BatchJob pricing bills Batch API inference at 60% of the standard price, and Batch covers kimi-k2.6 and kimi-k2.7-code only. This page stores the standard online price.
Q · 07 Are taxes or regional surcharges included? +
No. The Kimi pricing page says listed prices exclude applicable taxes and that checkout applies tax based on jurisdiction. AI//COST stores the vendor USD token price before taxes.
Q · 08 How accurate is the tokenizer estimate? +
The browser widget uses a moonshot-tokenizer-estimate chars-per-token estimate for English text. Actual billing comes from Kimi API usage fields and can differ for Chinese, code, images, video, or mixed-language prompts.
Y
Reviewed by Yaroslav Vikhariev Founder · AI//COST · Prices checked by hand against platform.kimi.ai · Last verified Oct 28, 2026
Methodology Report a correction More by Y.V.