InfoCapability
Together AI deprecates Qwen/Llama models on serverless — migrate to Gemma
AI Impact Summary
Together AI has deprecated several serverless models, including Qwen models and some Meta Llama models, effective April 15, 2026. This change impacts any applications currently utilizing these models for inference, requiring immediate migration planning to supported alternatives like Gemma or other available models. The shift to prepaid billing and dynamic rate limits also introduces new cost management considerations for users.
Together AI deprecates Qwen/Llama models on serverless — migrate to Gemma
Affected Systems
Qwen/Qwen3-Coder-Next-FP8deepseek-ai/DeepSeek-R1
- Date
- Date not specified
- Change type
- capability
- Severity
- info