AI Model Watch
Models / Moonshot

Kimi K2.6

GA

Superseded as flagship by kimi-k3 (confirmed 2026-09-02 — Moonshot's model list calls K3 "Kimi's most capable model to date"); K2.6 remains listed as a current multimodal model. Native multimodal (text, image, video input), supports thinking and non-thinking modes. Cache-miss input $0.95/M, cache-hit input $0.16/M, output $4.00/M (official pricing page, fetched 2026-06-28). Was the designated replacement for the deprecated kimi-k2 series and the retired kimi-latest/kimi-thinking-preview rows until 2026-09-02, when Moonshot re-pointed every one of those migration paths at kimi-k3. Release date 2026-04-20 from secondary sources (Yicai Global, kimi-k2.org, miraflow) - official page does not list release date or knowledge cutoff. Max output tokens not published. Moonshot consolidated its three per-model pricing pages into one table on or before 2026-09-11: chat-k3.md, chat-k26.md and chat-k27-code.md now all 307-redirect (temporary, per the header) to /docs/pricing/chat.md, which restates these exact rates and context window unchanged. Citation repointed there; our own link maintenance, not a provider price change, so no changelog event.

Kimi K2.6 by Moonshot costs $0.95 per 1M input tokens and $4 per 1M output tokens ($1.71/1M blended), with a 262K (262.144-token) context window. It is generally available (GA). Its weights are publicly downloadable, so it can be self-hosted on your own infrastructure.

Last verified: 26 Sep 2026 · sourced from official provider documentation

Provider
Moonshot
Status
GA
Input price
$0.95 / 1M tokens
Output price
$4 / 1M tokens
Cached input
$0.16 / 1M tokens
Blended price
$1.71 / 1M tokens
Context window
262.144 tokens (262K)
Max output
—
Modality
text, image, video
Open weights
Yes — self-hostable
Knowledge cutoff
—
Released
20 Apr 2026
API string
kimi-k2.6

Source: Moonshot official documentation ↗

See also

Compare Kimi K2.6 head-to-head

Kimi K2.6 vs Kimi K2.7 Code→

$1.71 vs $1.71 blended /M · same family

Kimi K2.6 vs Kimi K2.7 Code HighSpeed→

$1.71 vs $3.42 blended /M · same family

Kimi K2.6 vs Kimi K3→

$1.71 vs $6 blended /M · same family

FAQ

Kimi K2.6 — questions & answers

How much does Kimi K2.6 cost?

Kimi K2.6 is priced at $0.95 per 1M input tokens and $4 per 1M output tokens ($1.71/1M blended at a 3:1 input-to-output ratio), with cached input at $0.16 per 1M tokens.

What is the context window of Kimi K2.6?

Kimi K2.6 has a 262.144-token context window (262K).

Is Kimi K2.6 deprecated?

No — Kimi K2.6 is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.

Does Kimi K2.6 support image or vision input?

Yes — Kimi K2.6 accepts image input. Listed modalities: text, image, video.

Can I self-host Kimi K2.6?

Yes — Kimi K2.6 is an open-weight model: its weights are publicly downloadable, so you can run it on your own infrastructure (subject to its license) instead of only calling Moonshot's API.

What is the API model string for Kimi K2.6?

The API model identifier for Kimi K2.6 is "kimi-k2.6" when calling Moonshot's API.