AI Model Training

C-Eval C-Eval

A multi-level and multi-disciplinary assessment package for Chinese large models, containing 13948 questions, covering 52 subjects

CMMLU CMMLU

Comprehensive Chinese large model evaluation benchmark, covering multi-domain knowledge, measuring the knowledge and reasoning of the model in the Chinese context

Gemma 4 12B Google

Google's open source coder-less unified architecture multi-modal large model, the local running performance of 12B parameters exceeds the previous generation 27B

Giskard Giskard

Automated red team testing, quality assessment and enterprise-level monitoring platform for LLM application RAG and AI Agent

Core API AIStartMap

Cohere API provides full-stack enterprise-level large model services from Command generation to Embed/Rerank retrieval, supporting VPC private deployment and model fine-tuning.

Cosmos 3 NVIDIA

An open Physical AI world base model that integrates visual reasoning, world generation and action generation into the same model stack

Gradient AI Gradient

Cloud GPU infrastructure platform for AI training and inference, providing elastic computing and model deployment

Alibaba Cloud Bailian 百炼

Alibaba Cloud's one-stop large model development and application construction platform, covering model RAG, Agent and deployment

ACE-Ego ACE-Ego

One brain, multiple types of embodied operation VLA model research project

Databricks Databricks

Unify the data lake warehouse and AI platform, covering all scenarios of data engineering ML and large model training

Deepnote Deepnote

Cloud notebook collaboration platform for data teams, emphasizing real-time collaboration and production readiness

Agent Lightning Microsoft

Microsoft's agent training framework uses reinforcement learning and other methods to allow Agents to continuously optimize from experience.