MON, AUGUST 03, 2026
Independent · In‑Depth · Practitioner‑Tested
✎ Large Language Models

Kimi API Sunset August 31 — kimi-k2.5 and moonshot-v1 Discontinued, Migrate to kimi-k3 Now

Moonshot confirmed: kimi-k2.5 and all moonshot-v1 strings (8k/32k/128k/auto) discontinued August 31, 2026 — already blocked for new API users since K3 launch July 16. Migrate to kimi-k3 for general tasks or kimi-for-coding for coding workloads. K2.5 open weights (Hugging Face, Modified MIT) unaffected. August 31 is also the Claude Sonnet 5 pricing deadline. US Treasury sanctions warning on Moonshot still unresolved.

By AIToolsRecap August 3, 2026 5 min read 17 views
Home Articles Large Language Models Kimi Kimi API Legacy Sunset August 31: kimi-k2.5 and...

KIMI API DEPRECATION — ACTION REQUIRED BY AUGUST 31, 2026

Sunset date: August 31, 2026 — full platform discontinuation
Models being discontinued: kimi-k2.5 · moonshot-v1-8k · moonshot-v1-32k · moonshot-v1-128k · moonshot-v1-auto
Already unavailable to: Newly registered Kimi API users (since K3 launch July 16)
Recommended migration: kimi-k3 (flagship) · kimi-k2.7-code / kimi-for-coding (coding workloads)
Also already discontinued: kimi-k2 series (May 25, 2026) · kimi-latest (January 28, 2026)
K2.5 open weights: Still available under Modified MIT license on Hugging Face — self-hosting unaffected
August 31 collision: Same day as Claude Sonnet 5 intro pricing end ($2→$3/M)

What Is Being Discontinued and When

According to Moonshot's official API model documentation, kimi-k2.5 and the moonshot-v1 model series are no longer available to newly registered users following the Kimi K3 launch on July 16, 2026. A full platform sunset — meaning the strings stop responding for all users including existing ones — is scheduled for August 31, 2026. As the independent Kimi K2.5 model guide documents, the moonshot-v1 family has five variants — moonshot-v1-8k, moonshot-v1-32k, moonshot-v1-128k, and moonshot-v1-auto — all of which are included in the August 31 discontinuation. The only difference between the moonshot-v1 variants was their maximum context length. There are no capability differences between them.

This follows an aggressive deprecation schedule from Moonshot. The kimi-k2 series was discontinued on May 25, 2026. The kimi-latest slug was discontinued on January 28, 2026. Moonshot is explicitly consolidating its API lineup around three strings: kimi-k3 for flagship work, kimi-k2.6 for lower-cost general tasks, and kimi-for-coding (kimi-k2.7-code) for coding-specific workloads. Per Tosea.ai's K3 migration guide, Moonshot is "pruning the model lineup aggressively" and wants all API users on these three strings before the August 31 cut-off.

Migration Map — Which String to Use

If you useDeadlineMigrate toPrice
kimi-k2.5August 31, 2026kimi-k3 or kimi-k2.6$3/M or lower
moonshot-v1-8k/32k/128k/autoAugust 31, 2026kimi-k3$3/M
kimi-k2 seriesAlready discontinued (May 25)kimi-k3 or kimi-for-coding$3/M or lower
kimi-latestAlready discontinued (Jan 28)kimi-k3$3/M

The K2.5 Open Weights Are Not Affected

The Moonshot-hosted API for kimi-k2.5 is being discontinued. The open weights are not. As the K2.5 model page confirms, the separately published K2.5 weights remain available under the repository's Modified MIT License on Hugging Face. Teams that self-host K2.5 via their own inference infrastructure are unaffected by the August 31 API sunset. The sunset only affects calls made to Moonshot's hosted API endpoints.

August 31: Two Deprecation Deadlines on One Day

August 31, 2026 is now a dual deadline for API teams using either Kimi or Anthropic models. On the same day: kimi-k2.5 and moonshot-v1 series stop responding (Moonshot), and Claude Sonnet 5 intro pricing ends — the price moves from $2/$10/M to $3/$15/M and the new tokenizer adds up to 35% more tokens per equivalent text request (Anthropic). Teams that run mixed Kimi and Anthropic workloads should treat August 31 as a single planning event: test Kimi migrations to kimi-k3, measure the Sonnet 5 tokenizer impact on their content, and enable Anthropic prompt caching before both deadlines hit simultaneously.

Kimi: run this now to find affected calls
Search your codebase for: moonshot-v1, kimi-k2.5, kimi-latest. Replace all with kimi-k3 for general tasks or kimi-for-coding for coding workloads. Test on a staging environment before August 31.

Note on the US Treasury sanctions warning
The Treasury's July 2026 warning about potential Moonshot AI Entity List designation remains unresolved as of August 3. If you are an enterprise team expanding Kimi API commitments as part of the migration, verify current legal status with counsel before committing new production volume to the Moonshot API. The open-weight self-hosting path is lower-risk for enterprise use.

Sources: Kimi API official model list · K2.5 sunset details · Tosea.ai K3 migration guide · Kingy.ai K3 deep dive · Related: Kimi K3 open weights guide → · Sonnet 5 August 31 pricing deadline → · DeepSeek V4 Flash 0731 — migrate from deepseek-chat by October 24 →

Tags
KimiAI NewsGenerative AI2026

Spot an inaccuracy?

We verify facts before publishing and correct errors promptly. If something in this article is wrong or outdated, let us know.

Report an error →
💡 Kimi prompts
Prompt Guide
Best Kimi K3 Prompts for Coding (2026)
Kimi K3 ranks #1 globally on Design Arena frontend coding (1679 Elo) and #1 on SWE Marathon (42.0%) — making it the strongest available model for frontend development and long-horizon agentic coding tasks. At $3/$15/M with 1M context on Allegretto+ tier, it offers the best price-to-coding-capability ratio of any frontier model. These prompts are optimised for Kimi Code CLI, the Kimi API, and kimi.com on Allegretto+ or Vivace tiers.
Get Prompts →
Prompt Guide
Best Kimi K3 Prompts for Research (2026)
Kimi K3's 1M context window on Allegretto+ tier makes it one of the strongest research models available — it can ingest entire research papers, technical documentation sets, and large code repositories in a single session. At $0.30/M cached input, repeated context (like a standing research brief or knowledge base) becomes extremely cost-efficient. These prompts are optimised for kimi.com on Allegretto+ or Vivace tiers and the Kimi API.
Get Prompts →
Prompt Guide
Best Kimi K3 Prompts for Writing (2026)
Kimi K3 on Allegretto+ tier offers 1M context for writing tasks that require deep document awareness — editing a long manuscript, maintaining style consistency across a large content set, or rewriting with full context of everything written before. At $0.30/M cached input, repeatedly referencing a style guide or brand voice document costs almost nothing. These prompts are optimised for kimi.com on Allegretto+ and the Kimi API.
Get Prompts →