AI Model Catalogue

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...

Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...

gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...

Mistral Small 4.

Official mistral-embed-2312 Mistral AI model

Our cutting-edge language model for coding released August 2025.

Kimi K2 Thinking is the latest, most capable version of an open-source thinking model.

gemini-3-flash-preview — available to run locally via Ollama on CPU and GPU hardware.

minimax-m2 — available to run locally via Ollama on CPU and GPU hardware.

qwen3-coder-next — available to run locally via Ollama on CPU and GPU hardware.

Open-source Llama-3.1-70B-Instruct model from meta-llama — available for download and self-hosting on Hugging Face.

kimi-k2-thinking — available to run locally via Ollama on CPU and GPU hardware.

Open-source Llama-3.2-1B-Instruct model from meta-llama — available for download and self-hosting on Hugging Face.

Open-source Llama-3.1-8B-Instruct model from meta-llama — available for download and self-hosting on Hugging Face.

Open-source Qwen3-30B-A3B-Instruct-2507 model from qwen — available for download and self-hosting on Hugging Face.

Open-source Qwen3-Coder-30B-A3B-Instruct model from qwen — available for download and self-hosting on Hugging Face.

Open-source Qwen3-30B-A3B model from qwen — available for download and self-hosting on Hugging Face.

Open-source Qwen3-14B model from qwen — available for download and self-hosting on Hugging Face.

Open-source Qwen3-32B model from qwen — available for download and self-hosting on Hugging Face.

Open-source Qwen3-8B model from qwen — available for download and self-hosting on Hugging Face.

AI21 Jamba 1.6 Mini is a lightweight Mamba-Transformer hybrid optimized for cost-effective, high-throughput inference with an impressive 256K context window. An excellent choice for document-heavy workloads on a budget.

AI21 Jamba 1.6 Large uses a hybrid Mamba-Transformer architecture offering low memory footprint and high throughput compared to equivalent Transformer models. Features 256K context at a fraction of the inference cost.

Microsoft Phi-4 Mini is a 3.8B parameter compact model from Microsoft. Delivers impressive reasoning capabilities for edge and mobile deployment scenarios, with strong performance on math and coding tasks relative to its size.

IBM Granite 3.0 2B Instruct is an ultra-compact enterprise model excelling at summarization, extraction, and classification. The smallest model in the Granite family, suitable for edge deployments and constrained environments.

Amazon Nova Lite is a very low-cost multimodal model that can process image, video, and text inputs. Fast and accurate for a wide range of tasks requiring visual and language understanding.

Amazon Nova Micro is the fastest and most cost-effective text-only model in the Nova family, optimized for speed and low latency. Ideal for customer service, summarization, and translation at scale.

GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks. Training data up to Sep 2021.

One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge

A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge

An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory. Meant for use in roleplay/narrative situations.

A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.

Your prompt will be processed by a meta-model and routed to one of dozens of models (see below), optimizing for the best possible output. To see which model was used,...

Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use. Incorporates 8 experts (feed-forward networks) for a total of 47 billion...

GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks. Training data up to Sep 2021.

Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance. See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-haiku) #multimodal

WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model. It demonstrates highly competitive performance compared to leading proprietary models, and it consistently outperforms all existing state-of-the-art opensource models. It is...