Kimi K2.6
No shutdown before this date. Kimi K2.6 stays on Azure AI Foundry until at least Apr 16, 2027. Moonshot AI names GLM-5.3 as the replacement.
≥ 6 mo- Earliest retirement
- ≥ 184 daysApr 16, 2027 · Azure AI Foundry
- Arena Text
- 1,461#31 of 128 · Arena
- GPQA Diamond
- 90.8%#30 of 108 · Epoch AI
- Context window
- 262.1Ktokens
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Azure AI FoundryMicrosoft: Model retirement schedule - Microsoft Foundry (opens in a new tab) · checked Oct 14, 2026 | Kimi-K2.6:2026-04-20Preview | – | ≥ 6 moNot sooner than | – |
| Together AITogether AI: Changelog (opens in a new tab) · checked Oct 14, 2026 | moonshotai/Kimi-K2.6Shut down | – | Shut down | – |
| Fireworks AIFireworks AI: Kimi K2.6 model page (opens in a new tab) · checked Oct 14, 2026 | accounts/fireworks/models/kimi-k2p6Shut down | Shut down | GLM-5.3 |
Preview dates on Azure are a floor. Preview deployments are upgraded to a newer version or retired with at least 30 days' notice.
Fireworks also named Kimi K3 as a migration target.
Removed from serverless only; dedicated deployments of this model are unaffected.
What breaks when you move off Kimi K2.6
- Availability
- Availability
- Availability
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,461 ± 2 | #31 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 90.8% ± 1.7% | #30 of 108 | May 1, 2026 |
| Arena WebDev | Arena | 1,509 ± 4 | #30 of 69 | Oct 8, 2026 |
| FrontierMath Tier 4 | Epoch AI | 25.6% ± 7.1% | #35 of 58 | Jun 10, 2026 |
| FrontierMath Tiers 1-3 | Epoch AI | 57.2% ± 2.9% | #39 of 71 | Jun 10, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 96.1% ± 2.4% | #27 of 105 | May 2, 2026 |
| SimpleQA Verified | Epoch AI | 34.9% ± 1.5% | #45 of 68 | Aug 10, 2026 |
| SWE-bench Verified | Epoch AI | 76.7% ± 1.9% | #9 of 28 | May 8, 2026 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
Change history
ShutdownFireworks removes older DeepSeek, GLM, Muse and Kimi models from serverlessFireworks AI1 break
DeepSeek V4 Pro (0813), DeepSeek V4 Flash (0731), DeepSeek V4 Flash Vision Exp, Muse Glimmer 30B, Kimi K2.6, Kimi K2.7 Code and GLM 5.2 left serverless, including their Fast and US-only endpoints. Fireworks named DeepSeek V4.1 Flash, NVIDIA Nemotron 3.5 Lightning 30B A3B, GLM 5.3 and Kimi K3 as migration targets.
- Platform
- Fireworks AI
- Models
- DeepSeek-V4-Pro, DeepSeek-V4-Flash, DeepSeek-V4-Flash-Vision-Exp, Muse Glimmer 30B, Kimi K2.6, Kimi K2.7 Code, GLM-5.2
- Announced
- AvailabilityServerless requests to these model IDs stop working after the date. Dedicated deployments are unaffected.
ShutdownFireworks removes MiniMax M2.7, GPT OSS 20B, DeepSeek V4 Pro and fast Kimi variants from serverlessFireworks AI1 break
MiniMax M2.7, GPT OSS 20B and the original DeepSeek V4 Pro left serverless, along with the Turbo and Fast variants of Kimi K2.6 and the Fast variant of Kimi K2.7 Code. Fireworks named MiniMax M3, GPT OSS 120B, DeepSeek V4 Pro (0813) and the standard Kimi modes as migration targets.
- Platform
- Fireworks AI
- AvailabilityServerless requests to these model IDs stop working after the date. Dedicated deployments are unaffected.
ShutdownKimi K2.6 leaves Together serverlessTogether AI1 break
Together removed moonshotai/Kimi-K2.6 from serverless inference.
- Platform
- Together AI
- Model
- Kimi K2.6
- AvailabilityRequests to the removed serverless model ID fail. The model can still run on an on-demand dedicated endpoint, billed per minute.
Source: Together AI, Deprecations (opens in a new tab); Together AI, Changelog (opens in a new tab)