---
title: "Together AI platform release notes and changelog, January–June 2026 — Cloud Changelog"
description: "Together AI platform release notes from January–June 2026: 74 announcements. Latest: Qwen/Qwen3.5-397B-A17B deprecated and removed from serverless"
url: "https://cloudchangelog.com/products/ai/together-ai/2026-h1"
updated: "2026-11-18"
---

1. [Home](https://cloudchangelog.com/)
2. ›[AI providers](https://cloudchangelog.com/products/ai)
3. ›[Together AI platform](https://cloudchangelog.com/products/ai/together-ai)
4. ›January–June 2026

# Together AI platform release notes and changelog, January–June 2026

74 announcements, January–June 2026

[All Together AI platform changes](https://cloudchangelog.com/products/ai/together-ai) · 278 announcements since July 2025 · 3 GA · 2 deprecations

Free. Unsubscribe any time.

[RSS feed for Together AI platform](https://cloudchangelog.com/api/3LKjxEzy7iiSV3FYehOlOemjcXv/news/rss?provider=together&product=together-ai)

## June 202624

- Jun 29

  [Qwen/Qwen3.5-397B-A17B deprecated and removed from serverless](https://cloudchangelog.com/ai/together-2026-06-29-13484e91)
- Jun 26

  [GLM-5.1 and four other models deprecated on serverless](https://cloudchangelog.com/ai/together-2026-06-26-45a36547)
- Jun 25

  [Seedance 2.0 adds 4k resolution tier (up to 3840x2160) at $0.836 per second](https://cloudchangelog.com/ai/together-2026-06-25-aebc8406)
- Jun 24

  [Fine-tuning now supports Kimi-K2.7-Code and Kimi-K2.6](https://cloudchangelog.com/ai/together-2026-06-24-2bc2526c)
- Jun 24

  [GPU clusters add passive health checks and automatic node repair](https://cloudchangelog.com/ai/together-2026-06-24-71046a2d)
- Jun 24

  [POST /fine-tunes/estimate-price returns estimated fine-tuning job cost before launch](https://cloudchangelog.com/ai/together-2026-06-24-b7ed70ac)
- Jun 23

  [Fine-tuning jobs support early stopping based on validation loss](https://cloudchangelog.com/ai/together-2026-06-23-452ae08b)
- Jun 23

  [GET /whoami shows the API key, project and organization behind a request](https://cloudchangelog.com/ai/together-2026-06-23-c22e9b9e)
- Jun 23

  [Binary audio uploads for transcription and translation capped at 80 MB](https://cloudchangelog.com/ai/together-2026-06-23-cfd50f8c)
- Jun 22

  [Multiple LoRA adapters can share one dedicated endpoint, in preview](https://cloudchangelog.com/ai/together-2026-06-22-a5135300)
- Jun 22

  [GLM-5 deprecated on serverless, replaced by GLM-5.2](https://cloudchangelog.com/ai/together-2026-06-22-ad31c3d8)
- Jun 17

  [GLM-5.2 available on serverless with 262K context at $1.40 input per 1M tokens](https://cloudchangelog.com/ai/together-2026-06-17-483d31a2)
- Jun 17

  [Organization and project roles renamed to Admin, Developer and Editor labels](https://cloudchangelog.com/ai/together-2026-06-17-9eb573c2)
- Jun 15

  [Qwen3.5-397B-A17B deprecated on serverless after June 29, 2026](https://cloudchangelog.com/ai/together-2026-06-15-a05c568d)
- Jun 13

  [Kimi-K2.7-Code available on serverless with 262,144 context, $0.95 input per 1M](https://cloudchangelog.com/ai/together-2026-06-13-59047c5e)
- Jun 12

  [MiniMax-M3 on serverless with 524,288-token context, $0.30 input per 1M tokens](https://cloudchangelog.com/ai/together-2026-06-12-e4e4cb4c)
- Jun 11

  [Voxtral-Mini-3B-2507 deprecated on serverless, still available as dedicated endpoint](https://cloudchangelog.com/ai/together-2026-06-11-e0be98ec)
- Jun 9

  [Cached input pricing for GLM-5.1 and Qwen3.5-397B-A17B, DeepSeek-V4-Pro price cut](https://cloudchangelog.com/ai/together-2026-06-09-ad55d9ba)
- Jun 8

  [Fine-tuning file uploads get server-side schema validation with processing\_status](https://cloudchangelog.com/ai/together-2026-06-08-2d6feebc)
- Jun 4

  [Qwen3-Coder-480B-A35B-Instruct-FP8 deprecated on serverless; MiniMax-M2.7 suggested](https://cloudchangelog.com/ai/together-2026-06-04-3b9d2f8e)
- Jun 1

  [GET /fine-tunes/{id}/metrics endpoint returns fine-tuning training metrics](https://cloudchangelog.com/ai/together-2026-06-01-12276df6)
- Jun 1

  [Billing docs updated for multiple payment methods and prepaid-only access](https://cloudchangelog.com/ai/together-2026-06-01-66a5eb75)
- Jun 1

  [GPU clusters support Slurm startup scripts for node startup and per-job hooks](https://cloudchangelog.com/ai/together-2026-06-01-c8fe93a7)
- Jun 1

  [Compare evaluator adds disable\_position\_bias\_correction, halving judge cost and latency](https://cloudchangelog.com/ai/together-2026-06-01-da917db6)

## May 202618

- May 29

  [Price increases for Qwen3.5-9B, Llama 3 8B Lite and Llama 3.3 70B Turbo](https://cloudchangelog.com/ai/together-2026-05-29-44f57935)
- May 27

  [FLUX.1-krea-dev deprecated and no longer available on serverless](https://cloudchangelog.com/ai/together-2026-05-27-7cf8c0a6)
- May 25

  [Gemma, Qwen2.5 and Llama 3.2 among 17 models added to dedicated endpoints](https://cloudchangelog.com/ai/together-2026-05-25-2dba6c36)
- May 25

  [Quickstart added for Seedance 2.0 video generation, covering text-to-video and editing](https://cloudchangelog.com/ai/together-2026-05-25-329f0547)
- May 25

  [Seedream 5.0 Lite, HappyHorse 1.0 and Veo 3.1 models now on serverless](https://cloudchangelog.com/ai/together-2026-05-25-4ebc7fdf)
- May 22

  [Qwen3.7-Max available on serverless at $2.50 input / $7.50 output per 1M tokens](https://cloudchangelog.com/ai/together-2026-05-22-38361561)
- May 22

  [GPU clusters support external OIDC authentication for Kubernetes clusters](https://cloudchangelog.com/ai/together-2026-05-22-57e43818)
- May 21

  [Kimi-K2.5 deprecated and removed from serverless](https://cloudchangelog.com/ai/together-2026-05-21-88bb1360)
- May 15

  [pearl-ai/gemma-4-31b-it available on serverless with 32,000 context length](https://cloudchangelog.com/ai/together-2026-05-15-a90f36e5)
- May 14

  [DeepSeek-R1, DeepSeek-V3.1 and Qwen3-Coder-Next-FP8 deprecated on serverless](https://cloudchangelog.com/ai/together-2026-05-14-93e13a9b)
- May 14

  [google/gemma-4-31b-it price rises to $0.39 input, $0.97 output from May 21, 2026](https://cloudchangelog.com/ai/together-2026-05-14-d15b76cd)
- May 8

  [External collaborators can be invited to projects, in beta](https://cloudchangelog.com/ai/together-2026-05-08-147d21a1)
- May 8

  [HappyHorse 1.0 t2v and Seedance 2.0 video models available on serverless](https://cloudchangelog.com/ai/together-2026-05-08-904ebe60)
- May 7

  [Audio transcription and translation endpoints accept .ogg, .opus and .aac files](https://cloudchangelog.com/ai/together-2026-05-07-150efc6f)
- May 7

  [Together CLI uses tg as the canonical command name, with a refreshed command tree](https://cloudchangelog.com/ai/together-2026-05-07-32dfb3b2)
- May 7

  [task field no longer required in audio verbose JSON transcription and translation responses](https://cloudchangelog.com/ai/together-2026-05-07-bacd433a)
- May 6

  [Slurm GPU clusters move to Slurm-on-Kubernetes stack with self-healing workers](https://cloudchangelog.com/ai/together-2026-05-06-d3b61bba)
- May 1

  [MiniMax-M2.5 deprecated and removed from serverless](https://cloudchangelog.com/ai/together-2026-05-01-509ecdff)

## April 202613

- Apr 30

  [TTS requests add pronunciation\_dict parameter for custom pronunciation rules](https://cloudchangelog.com/ai/together-2026-04-30-2239eeda)
- Apr 30

  [Deployments can autoscale on any custom Prometheus metric from /metrics](https://cloudchangelog.com/ai/together-2026-04-30-c00d27d0)
- Apr 28

  [Fine-tuning adds Qwen3.6-35B-A3B, Gemma 4 31B and Gemma 4 26B-A4B](https://cloudchangelog.com/ai/together-2026-04-28-bc9f8387)
- Apr 24

  [Serverless adds Cogito v2.1 671B, Vidu Q3 and Q3 Turbo, and Wan 2.7 video models](https://cloudchangelog.com/ai/together-2026-04-24-05478ed2)
- Apr 24

  [DeepSeek-V4-Pro added to serverless with 512,000-token context, $2.10 input per 1M](https://cloudchangelog.com/ai/together-2026-04-24-3f74b1b1)
- Apr 24

  [No-packing fine-tuning jobs now priced on len(dataset) max\_seq\_length tokens](https://cloudchangelog.com/ai/together-2026-04-24-f8313fe3)
- Apr 22

  [Kimi-K2.6 from Moonshot AI now available on serverless](https://cloudchangelog.com/ai/together-2026-04-22-4767888e)
- Apr 22

  [Build, Scale and Enterprise tiers retired; billing fully prepaid with dynamic rate limits](https://cloudchangelog.com/ai/together-2026-04-22-c0c8801c)
- Apr 15

  [google/gemma-3n-E4B-it pricing rises to $0.06 input, $0.12 output per 1M tokens](https://cloudchangelog.com/ai/together-2026-04-15-fbd5fbeb)
- Apr 14

  [Qwen3-VL-8B-Instruct, Qwen3-235B-A22B-Thinking-2507 and Mixtral-8x7B deprecated](https://cloudchangelog.com/ai/together-2026-04-14-64dd8a34)
- Apr 11

  [MiniMax-M2.7 now available on serverless](https://cloudchangelog.com/ai/together-2026-04-11-a916efcc)
- Apr 8

  [Gemma 4 31B-it and GLM-5.1 available on serverless](https://cloudchangelog.com/ai/together-2026-04-08-1de8750f)
- Apr 2

  [GLM-4.5-Air-FP8 and GLM-4.7 deprecated, along with Qwen3-Next-80B-A3B-Instruct](https://cloudchangelog.com/ai/together-2026-04-02-60d7fe8d)

## March 20264

- Mar 31

  [Llama-4-Maverick-17B-128E-Instruct-FP8 deprecated](https://cloudchangelog.com/ai/together-2026-03-31-12978f25)
- Mar 10

  [MiniMax-M2.5 cached input tokens priced at $0.06 per 1M, 80% off](https://cloudchangelog.com/ai/together-2026-03-10-8895ba0a)
- Mar 7

  [Qwen3.5-9B available on serverless](https://cloudchangelog.com/ai/together-2026-03-07-341428d0)
- Mar 6

  [Kimi-K2-Thinking and Mxbai-Rerank-Large-V2 deprecated, with two other models](https://cloudchangelog.com/ai/together-2026-03-06-ca0affca)

## February 20269

- Feb 25

  [FLUX.1-dev, Qwen3-VL-32B-Instruct and 11 other models deprecated](https://cloudchangelog.com/ai/together-2026-02-25-fd6a2bf7)
- Feb 16

  [Qwen3.5-397B-A17B available on serverless](https://cloudchangelog.com/ai/together-2026-02-16-c9e8f7f5)
- Feb 15

  [MiniMax-M2.5 now available on serverless](https://cloudchangelog.com/ai/together-2026-02-15-e57d0017)
- Feb 13

  [GLM-5 now available on serverless](https://cloudchangelog.com/ai/together-2026-02-13-69d43a4e)
- Feb 12

  [Dedicated Container Inference (DCI), formerly BYOC, launches for custom models](https://cloudchangelog.com/ai/together-2026-02-12-2ee8a659)
- Feb 6

  [Ten models deprecated, including Llama-4-Scout-17B-16E and Qwen2.5-72B-Instruct-Turbo](https://cloudchangelog.com/ai/together-2026-02-06-29702306)
- Feb 4

  [Python SDK v2.0 released, type-safe and OpenAPI-generated, with beta instant clusters APIs](https://cloudchangelog.com/ai/together-2026-02-04-ea5e6705)
- Feb 3

  [Qwen3-Coder-Next-FP8 now available on serverless](https://cloudchangelog.com/ai/together-2026-02-03-09ec6a9d)
- Feb 3

  [DeepSeek-R1-0528-tput deprecated and no longer available](https://cloudchangelog.com/ai/together-2026-02-03-a24ba0df)

## January 20266

- Jan 29

  [Mistral-7B-Instruct-v0.3 and GLM-4.6 now redirect to Ministral-3-14B and GLM-4.7](https://cloudchangelog.com/ai/together-2026-01-29-00ee7420)
- Jan 27

  [Kimi-K2.5 now available on serverless](https://cloudchangelog.com/ai/together-2026-01-27-a82a3f68)
- Jan 23

  [DeepSeek-V3-0324 automatically redirected to DeepSeek-V3.1](https://cloudchangelog.com/ai/together-2026-01-23-6fccd77c)
- Jan 21

  [Prompt caching always on for new dedicated endpoints; disable\_prompt\_cache deprecated](https://cloudchangelog.com/ai/together-2026-01-21-1e77031e)
- Jan 9

  [GLM-4.7 now available on serverless](https://cloudchangelog.com/ai/together-2026-01-09-2193954d)
- Jan 5

  [Qwen2.5-VL-72B-Instruct deprecated and no longer available](https://cloudchangelog.com/ai/together-2026-01-05-95ae1a57)

## More Together AI platform history

- [Latest changes](https://cloudchangelog.com/products/ai/together-ai)
- [2025 · 27](https://cloudchangelog.com/products/ai/together-ai/2025)

## Products related to Together AI platform

- [OpenAI API](https://cloudchangelog.com/products/ai/openai)
- [Gemini API](https://cloudchangelog.com/products/ai/gemini-api)
- [Claude Platform](https://cloudchangelog.com/products/ai/anthropic)
- [xAI API](https://cloudchangelog.com/products/ai/xai)
- [Mistral](https://cloudchangelog.com/products/ai/mistral)
- [Cerebras Inference](https://cloudchangelog.com/products/ai/cerebras)

[All AI providers product changelogs](https://cloudchangelog.com/products/ai)

## Contact

- [YouTube](https://www.youtube.com/@cloudchangelog)

```json
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    {
      "@type": "ListItem",
      "name": "Home",
      "position": 1,
      "item": "https://cloudchangelog.com/"
    },
    {
      "@type": "ListItem",
      "name": "AI providers",
      "position": 2,
      "item": "https://cloudchangelog.com/products/ai"
    },
    {
      "@type": "ListItem",
      "name": "Together AI platform",
      "position": 3,
      "item": "https://cloudchangelog.com/products/ai/together-ai"
    },
    {
      "@type": "ListItem",
      "name": "January–June 2026",
      "position": 4
    }
  ]
}
```
