Releasesby IbisPost-training: MiniMax M3.1 Flash
UpdateyesterdayUnverifiedAdded Oct 24, 2026
MiniMaxAI/MiniMax-M3.1-Flash is available for post-training : LoRA training on B200-SXM , a 262,144-token context for both training and sampling, rank up to 32, and up to three adapter sessions per resource. The cookbook renders its prompts with the MiniMax codec. See Supported models .
More Together AI platform releases
Every Together AI platform releaseNew serverless models
The following models are now available on serverless : MiniMaxAI/MiniMax-M3.1-Flash : 1,048,576 context length, NVFP4 quantization, function calling, and structured outputs. Pricing: $0.15 input / $0.60 output / $0.015 cached input (per 1M tokens).
Together SDK and CLI 3.0
Together Python SDK 3.0, TypeScript SDK 1.0, and Together CLI 3.0 are now available.
What's new:
- Dedicated model inference and GPU clusters out of beta : Manage dedicated model inference and GPU clusters with client.endpoints , client.models , and client.clusters , or with tg endpoints , tg models , and tg clusters in the CLI.
New serverless models
The following models are now available on serverless : deepseek-ai/DeepSeek-V4.1-Flash-1017 : 1,000,000 context length, FP8 quantization, function calling, and structured outputs. Pricing: $0.30 input / $1.20 output / $0.006 cached input (per 1M tokens).
New serverless models
The following models are now available on serverless : reflection/Beam-1.0-501B-A23B : 1,048,576 context length, NVFP4 quantization, function calling, and structured outputs. Pricing: $0.40 input / $1.20 output / $0.04 cached input (per 1M tokens).
Geo-specific inference
You can now keep serverless inference processing in the United States or Canada. Set your client's base URL to https://us-api.together.ai/v1 for the United States or https://ca-api.together.ai/v1 for Canada, using your existing API key and model IDs. There are no GPUs to provision.
New serverless models
The following models are now available on serverless : meta-models/Muse-Spark-Open-816B-A42B : 1,048,576 context length, NVFP4 quantization, function calling, and structured outputs. Pricing: $1.25 input / $4.25 output / $0.15 cached input (per 1M tokens).
Post-training
Post-training lets you write your own training loop and run it against Together AI GPUs. Your code builds each batch and chooses the loss function and learning rate, while the GPUs you provision run the forward and backward passes, optimizer steps, and sampling. It bills per GPU-minute.
Also shipped on Oct 23, 2026
Together AI in October 2026Retrieve Cortex Search (Public Preview)
The CORTEX_SEARCH_RETRIEVE table function is now in public preview, allowing users to execute low-latency search queries directly from SQL for RAG and chatbot applications.
DataFrame caching on serverless compute is now in Public Preview
Databricks now supports DataFrame caching on serverless compute via df.cache() or df.persist(), allowing users to reuse results within a session.
Box Audit Logs managed ingestion connector (Beta)
The Box Audit Logs connector for Lakeflow Connect is now available in beta, enabling the ingestion of enterprise audit events into Databricks.
Manage Test Engine workflows with Terraform and the REST API
Users can now manage Test Engine workflows using the buildkite_test_suite_workflow Terraform resource or the REST API, enabling code-based configuration of monitors and actions.
Flagship - Target Flagship feature flags with structured evaluation context
Flagship now supports nested objects and arrays in evaluation context, allowing developers to target feature flags based on complex user profiles and tags.
Responses API for the Cortex REST API (Preview)
The new Responses API in the Cortex REST API provides OpenAI-compatible endpoints for accessing models.