Qwen models

Amazon SageMaker AI now supports serverless model customization for Muse Glimmer, Qwen 3.8 and Ministral 3 models

AWS · Qwen3.8 27B

Full fine-tuning max batch size for Qwen 27B models lowered to 4

Together AI · Qwen3.8 27B · Qwen3.6 27B · Qwen3.5 27B

Kokoro-82M and Qwen3-TTS-12Hz-1.7B-CustomVoice on dedicated endpoints

Together AI · Qwen3-TTS-12Hz-1.7B-CustomVoice

Qwen 27B models support 131,072-token SFT context, with lower batch size limits

Together AI · Qwen3.8 27B · Qwen3.6 27B · Qwen3.5 27B

qwen-3.8-27b Developer tier rate limit raised temporarily to 750K tokens per minute

Cerebras · Qwen3.8 27B

Qwen3.7-Max and Qwen3.8-Flash prices lowered, effective September 22, 2026

Together AI · Qwen3.7-Max · Qwen3.8-Flash

Qwen3.6-35B-A3B-NVFP4 and Wan2.1-T2V-1.3B-Diffusers models now available on Amazon SageMaker JumpStartGA

AWS · Qwen3.6-35B-A3B-NVFP4

Three models deprecated for fine-tuning, including Nemotron-3 and Qwen3.5-35B-A3B-Base

Together AI · Qwen3.5-35B-A3B-Base

qwen-3.8-27b available on public endpoints with image input and configurable reasoning

Cerebras · Qwen3.8 27B

Qwen3.8-Flash on serverless with 1M context, at $0.15 input / $0.47 output per 1M tokens

Together AI · Qwen3.8-Flash

DeepSeek V4 Flash 0731, Qwen 3.8 27B and Muse Glimmer 30B added for LoRA training

Fireworks AI · Qwen3.8 27B

Muse-Glimmer-30B and Qwen 3.8-27B models now available on Amazon SageMaker JumpStartGA

AWS · Qwen3.8 27B

Qwen 3.5 9B and Qwen 3.6 27B deprecated from Serverless Training on August 26, 2026

Fireworks AI · Qwen3.5 9B · Qwen3.6 27B · Qwen3.8 27B

Qwen3.8-27B now available for fine-tuning

Together AI · Qwen3.8 27B

Fine-tuning training quality improves for Qwen3.5, Qwen3.6 and Nemotron 3 models

Together AI

FLUX.1-schnell, Qwen2.5-7B and Kimi-K2.6 deprecated on serverless

Together AI · Qwen2.5-7B

Workers AI - Qwen 3.8 27B now available on Workers AIGA

Cloudflare · Qwen3.8 27B

Qwen3.8-2.4T-A95B now available on serverless, in FP4 quantization

Together AI · Qwen3.8-2.4T-A95B

LocateAnything-3B, Qwen-AgentWorld-35B-A3B, and Qwen3.5-122B-A10B models now available on Amazon SageMaker JumpStart

AWS · Qwen-AgentWorld-35B-A3B · Qwen3.5-122B-A10B

Many Qwen, Kimi, GLM, DeepSeek and Llama models deprecated for fine-tuning

Together AI · Qwen3-235B-A22B

Qwen/Qwen3.6-27B now available for fine-tuning

Together AI · Qwen3.6 27B

Qwen3 embedding and reranking models for retrieval are now available in Amazon SageMaker JumpStart

AWS · Qwen3-VL-Embedding-2B · Qwen3-Reranker-4B

Qwen3-235B-A22B-Instruct-2507-tput, Llama-3-8B-Instruct-Lite and GLM-5.1 deprecated on serverless

Together AI · Qwen3-235B-A22B-Instruct-2507-tput

Serverless adds Gemini 3.1 Flash-Lite Image, Qwen3.6-35B-A3B-Lora and HappyHorse 1.1 video models

Together AI · Qwen3.6-35B-A3B-Lora

Qwen/Qwen3.5-397B-A17B deprecated and removed from serverless

Together AI · Qwen3.5-397B-A17B

Kimi K2.5 and Qwen 3.6 Plus deprecated from serverless

Fireworks AI · Qwen 3.6 Plus · Qwen 3.7 Plus

GLM-5.1 and four other models deprecated on serverless

Together AI · Qwen3-235B-A22B-Instruct-tput

Qwen3.5-397B-A17B deprecated on serverless after June 29, 2026

Together AI · Qwen3.5-397B-A17B

Kimi K2.7 Code, MiniMax M3 and Qwen 3.7 Plus added to the Model Library

Fireworks AI · Qwen 3.7 Plus

Cached input pricing for GLM-5.1 and Qwen3.5-397B-A17B, DeepSeek-V4-Pro price cut

Together AI · Qwen3.5-397B-A17B

Qwen3-Coder-480B-A35B-Instruct-FP8 deprecated on serverless; MiniMax-M2.7 suggested

Together AI · Qwen3-Coder-480B-A35B-Instruct-FP8

Price increases for Qwen3.5-9B, Llama 3 8B Lite and Llama 3.3 70B Turbo

Together AI · Qwen3.5 9B

Gemma, Qwen2.5 and Llama 3.2 among 17 models added to dedicated endpoints

Together AI · Qwen2.5-32B

Qwen3.7-Max available on serverless at $2.50 input / $7.50 output per 1M tokens

Together AI · Qwen3.7-Max

SageMaker AI now supports serverless model customization for Qwen3.6

AWS · Qwen3.6 27B

New models for image generation and text embeddings are now available in Amazon SageMaker JumpStart

AWS · Qwen3-Embedding-0.6B

Three new models for speech recognition and text-to-speech are now available in Amazon SageMaker JumpStart

AWS · Qwen3-TTS-12Hz-1.7B-CustomVoice · Qwen3-TTS-12Hz-1.7B-Base · Qwen3-ASR-1.7B

DeepSeek-R1, DeepSeek-V3.1 and Qwen3-Coder-Next-FP8 deprecated on serverless

Together AI · Qwen3-Coder-Next-FP8

4 new Qwen models for multimodal reasoning, agentic coding, and multilingual applications are now available in Amazon SageMaker JumpStart

AWS · Qwen3.5-27B-FP8 · Qwen3.6-35B-A3B · Qwen3.5-0.8B · Qwen3.5-2B

Fine-tuning adds Qwen3.6-35B-A3B, Gemma 4 31B and Gemma 4 26B-A4B

Together AI · Qwen3.6-35B-A3B

Amazon SageMaker AI now supports serverless model customization for Qwen3.5 models

AWS · Qwen3.5 4B · Qwen3.5 9B · Qwen3.5 27B

Five new Qwen models for coding agents and efficient reasoning are now available in Amazon SageMaker JumpStart

AWS · Qwen3-Coder-Next · Qwen3-30B-A3B · Qwen3-30B-A3B-Thinking · Qwen3-Coder-30B-A3B-Instruct

Qwen3-VL-8B-Instruct, Qwen3-235B-A22B-Thinking-2507 and Mixtral-8x7B deprecated

Together AI · Qwen3-VL-8B-Instruct · Qwen3-235B-A22B-Thinking

NVIDIA Nemotron-3-Super-120B, Qwen3.5-9B, and Qwen3.5-27B models now available on Amazon SageMaker JumpStart

AWS · Qwen3.5 9B · Qwen3.5 27B

AI Search - New Workers AI models for text generation and embedding in AI Search

Cloudflare · Qwen3-30B-A3B · Qwen3-Embedding-0.6B

GLM-4.5-Air-FP8 and GLM-4.7 deprecated, along with Qwen3-Next-80B-A3B-Instruct

Together AI · Qwen3-Next-80B-A3B-Instruct

Amazon SageMaker AI now supports serverless reinforcement fine-tuning for 12 additional models

AWS · Qwen2.5 72B Instruct

Qwen3.5-9B available on serverless

Together AI · Qwen3.5 9B

FLUX.1-dev, Qwen3-VL-32B-Instruct and 11 other models deprecated

Together AI · Qwen3-VL-32B-Instruct

Qwen3.5-397B-A17B available on serverless

Together AI · Qwen3.5-397B-A17B

Ten models deprecated, including Llama-4-Scout-17B-16E and Qwen2.5-72B-Instruct-Turbo

Together AI · Qwen2.5-72B-Instruct-Turbo · Qwen3-235B-A22B-fp8-tput

Fireworks adds video and audio inputs for Qwen3 Omni and Molmo2, and AWS S3 BYOB training data

Fireworks AI · Qwen3 Omni

Qwen3-Coder-Next-FP8 now available on serverless

Together AI · Qwen3-Coder-Next-FP8

DeepSeek OCR, MiniMax M2.1, and Qwen3-VL-8B-Instruct models are now available on SageMaker JumpStartGA

AWS · Qwen3-VL-8B-Instruct

Qwen2.5-VL-72B-Instruct deprecated and no longer available

Together AI · Qwen 2.5 VL 72B Instruct

DeepSeek V3.2 on serverless, plus cached and uncached token pricing in Model Library

Fireworks AI · Qwen3-VL-32B-Instruct · Qwen3-VL-8B-Instruct

qwen-3-coder-480b deprecated on Cerebras Inference, migrate to Z.ai GLM 4.7

Cerebras · Qwen 3 Coder 480B

Qwen3 models are now available fully managed in Amazon BedrockGA

AWS · Qwen3-Coder-480B-A35B-Instruct · Qwen3-Coder-30B-A3B-Instruct · Qwen3-235B-A22B-Instruct · Qwen3-32B

Qwen3-Next-80B-A3B Thinking and Instruct models available

Together AI · Qwen3-Next-80B-A3B-Thinking · Qwen3-Next-80B-A3B-Instruct

Fine-tuning context lengths increased for DeepSeek-R1-Distill, Gemma 3, Qwen and Llama models

Together AI · Qwen3-8B

17 models deprecated, including Llama 3.2 Vision and FLUX.1 canny, depth and redux

Together AI · Qwen2.5-14B

Cerebras model deprecated, migration to Qwen 3 32B recommended

Cerebras · Qwen3-32B

Supervised fine-tuning supports Qwen3 MoE and GPT OSS models

Fireworks AI

Qwen 3 Coder 480B (qwen-3-coder-480b) in preview on Cerebras Inference

Cerebras · Qwen 3 Coder 480B

Qwen 3 235B Thinking (qwen-3-235b-a22b-thinking-2507) in preview on Cerebras Inference

Cerebras · Qwen 3 235B Thinking

Migrate to Qwen 3 235B Instruct or Thinking (coming soon) as deprecation notice

Cerebras · Qwen 3 235B Instruct · Qwen 3 235B Thinking

Qwen 3 235B Instruct in preview as qwen-3-235b-a22b-instruct-2507

Cerebras · Qwen 3 235B Instruct

Fine-tuning for Qwen 2.5 VL vision-language models, with up to 64K context

Fireworks AI · Qwen 2.5 VL 3B Instruct · Qwen 2.5 VL 7B Instruct · Qwen 2.5 VL 32B Instruct · Qwen 2.5 VL 72B Instruct

Qwen3-Coder-480B-A35B-Instruct-FP8 added, with a 256K context length

Together AI · Qwen3-Coder-480B-A35B-Instruct-FP8

Qwen 3 235B added as qwen-3-235b-a22b

Cerebras · Qwen 3 235B

Amazon Bedrock Custom Model Import now supports Qwen modelsGA

AWS · Qwen 2.5 Coder · Qwen 2.5 VL

Qwen 3 32B added, with multi-turn tool call support

Cerebras · Qwen3-32B