AI Language Model Library

The newest series of Code-Specific Qwen models features major enhancements in code generation, reasoning, and correction. These advancements lead to more effective coding solutions.

Solar Pro Preview is an advanced large language model (LLM) featuring 22 billion parameters, optimized for deployment on a single GPU.

The Qwen2.5 models have been pretrained on Alibaba's latest extensive dataset, which includes up to 18 trillion tokens. This model can handle up to 128K tokens and offers multilingual support.

Mistral Small is a lightweight model designed for cost-effective use in tasks like translation and summarization.

A series of multimodal LLMs (MLLMs) designed for vision-language understanding.

An upgraded version of DeekSeek-V2 that integrates the general and coding abilities of both DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct.

Yi-Coder is a series of open-source code language models that delivers state-of-the-art coding performance with fewer than 10 billion parameters.

SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters.

The IBM Granite Guardian 3.0 2B and 8B models are designed to detect risks in prompts and/or responses.

Cohere For AI's language models trained to perform well across 23 different languages.

The IBM Granite 1B and 3B models are the first mixture of experts (MoE) Granite models from IBM designed for low latency usage.

The IBM Granite 2B and 8B models are designed to support tool-based use cases and support for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.

Llama-3.1-Nemotron-70B-Instruct is a large language model customized by NVIDIA to improve the helpfulness of LLM generated responses to user queries.

ShieldGemma is set of instruction tuned models for evaluating the safety of text prompt input and text output responses against a set of defined safety policies.

Llama Guard 3 is a series of models fine-tuned for content safety classification of LLM inputs and responses.

Meta's Llama 3.2 goes small with 1B and 3B models.

A lightweight AI model consisting of 3.8 billion parameters surpasses the performance of similarly sized and larger models.

A cutting-edge 12B model featuring a 128k context length, developed by Mistral AI in partnership with NVIDIA.

Meta has released Llama 3.1, a cutting-edge model offered in parameter sizes of 8B, 70B, and 405B.

Mistral Large 2 is the latest flagship model from Mistral, offering enhanced performance in code generation, mathematics, and reasoning, thanks to its 128k context window and support for numerous languages. Its capabilities are significantly improved compared to previous models.

Qwen2 Math consists of a collection of specialized math language models based on the Qwen2 LLMs. These models greatly surpass the mathematical performance of both open-source and even some closed-source models, such as GPT4o.

Hermes 3 is the newest iteration of Nous Research's flagship series of LLMs. It continues the legacy of the Hermes series.

Google Gemma 2 is now offered in three sizes: 2B, 9B and 27B.

Meta Llama 3 is the most advanced openly accessible language model available so far. It stands out as the leading choice among open-source LLMs.

Qwen2 is a newly launched series of extensive language models developed by the Alibaba Group.

A code language model based on an open-source Mixture-of-Experts approach, it delivers performance similar to GPT-4 Turbo for code-related tasks.

Phi-3 consists of advanced, lightweight open models, including 3B (Mini) and 14B (Medium) versions, developed by Microsoft.

Cohere has launched Aya 23, a cutting-edge family of multilingual models that accommodate 23 different languages.

Mistral AI has released the 7B model, now updated to version 0.3.

Mistral AI has released a set of Mixture of Experts (MoE) models featuring open weights, available in parameter sizes of 8x7b and 8x22b.

CodeGemma comprises robust, lightweight models capable of various coding tasks, including code completion, generation, natural language understanding, mathematical reasoning, and instruction adherence. These models are designed to efficiently handle diverse programming challenges.

Command R is a Large Language Model designed for effective conversational interaction and handling extended context tasks. It is optimized to facilitate engaging dialogue and manage lengthy discussions.

Command R+ is a robust and scalable large language model designed specifically for optimal performance in retrieval augmented generation (RAG) and real-world enterprise applications.

LLaVA is a novel large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding. Updated to version 1.6.

Gemma is a cutting-edge family of lightweight open models developed by Google DeepMind. It has recently been updated to version 1.1.

Qwen 1.5 is a range of large language models developed by Alibaba Cloud, featuring parameter sizes from 0.5 billion to 110 billion.

Llama 2 consists of a series of foundational language models with parameter sizes between 7 billion and 70 billion.

A powerful language model capable of generating and discussing code based on text prompts. It leverages natural language to facilitate coding tasks.

Eric Hartford developed the uncensored 8x7b and 8x22b fine-tuned models using the Mixtral mixture of experts, which are particularly effective for coding tasks.

The Uncensored Llama 2 model was developed by George Sung and Jarrad Hope.

DeepSeek Coder is a powerful coding model that has been trained on two trillion tokens of code and natural language. Its extensive training allows it to perform well in various coding tasks.

Phi-2 is a 2.7 billion parameter language model developed by Microsoft Research, showcasing exceptional reasoning and language comprehension skills.

The uncensored Dolphin model, which is built on Mistral, performs exceptionally well in coding tasks. It has been updated to version 2.8.

A versatile model with a parameter range of 3 billion to 70 billion, designed for use on entry-level hardware.

NVIDIA has developed a small language model that is optimized for commercial use, particularly in roleplay, retrieval-augmented generation (RAG) question answering, and function calling. It is designed to be user-friendly for various applications.

Dolphin 2.9, developed by Eric Hartford and built on Llama 3, comes in 8B and 70B sizes. It boasts a range of capabilities in instruction, conversation, and coding.

A state-of-the-art fact-checking model developed by Bespoke Labs.

Mistral OpenOrca is a 7 billion parameter model that has been fine-tuned from the Mistral 7B model with the OpenOrca dataset.

StarCoder2 represents the next evolution of openly trained LLMs for code, available in three sizes: 3B, 7B, and 15B parameters.

A series of models that convert HTML content to Markdown content, which is useful for content conversion tasks.

Zephyr comprises refined versions of the Mistral and Mixtral models, designed to serve as effective assistants. These models are specifically trained to enhance their helpfulness.

Yi 1.5 is an advanced bilingual language model that delivers exceptional performance. It excels in understanding and generating text in two languages.

A Llama 2-based model has been fine-tuned to enhance its ability to engage in Chinese dialogue. This adjustment aims to improve conversational performance in that language.

A high-performing model trained with a new technique called Reflection-tuning that teaches a LLM to detect mistakes in its reasoning and correct course.

A LLaVA model, fine-tuned from Llama 3 Instruct, has achieved improved scores across multiple benchmarks.

A general-purpose chat model built on Llama and Llama 2, offering context sizes ranging from 2K to 16K.

Nous Research offers a robust family of models that excel in scientific discussions and coding tasks. These powerful models are designed to enhance productivity in various research applications.

Wizard Vicuna Uncensored is a model with 7B, 13B, and 30B parameters, developed from Llama 2 by Eric Hartford. It is an uncensored version of the original model.

The TinyLlama SLM (small language model) is a compact AI language model with 1.1B parameters, designed for local on-premise inference on consumer grade hardware.

Codestral is the inaugural code model from Mistral AI, specifically developed for tasks involving code generation.

StarCoder is a model designed for code generation, trained in over 80 programming languages.

Microsoft AI has developed a cutting-edge large language model that enhances performance in complex chat, multilingual interactions, reasoning, and agent applications.

An open-source model family trained on diverse data sets has outperformed ChatGPT on multiple benchmarks. It has been updated to version 3.5-0106.

Eric Hartford developed an experimental model with 1.1 billion parameters, utilizing the new Dolphin 2.8 dataset and based on TinyLlama.

OpenHermes 2.5 is a 7B model that has been fine-tuned by Teknium using Mistral and entirely open datasets.

Advanced code generation model.

Stable Code 3B is a coding model that offers instruct and code completion options comparable to larger models like Code Llama 7B, which is 2.5 times its size.

CodeQwen1.5 is a substantial language model that has been pretrained using extensive code datasets.

A well-tuned Mistral model that effectively covers both domain and language.

The model specializes in solving mathematical and logical challenges. It emphasizes analytical reasoning and problem-solving skills.

Stable LM 2 is an advanced language model with 1.6 billion and 12 billion parameters, designed using multilingual data in several languages including English, Spanish, German, Italian, French, Portuguese, and Dutch.

A code generation model utilizing Code Llama.

IBM has developed a family of open foundation models designed for code intelligence.

A 7B and 15B uncensored version of the Dolphin model family, which is highly effective at coding, is derived from StarCoder2.

Nous Research has developed general use models based on Llama and Llama 2.

SQLCoder is a code completion model specifically optimized for SQL generation tasks, built upon StarCoder. It offers enhanced capabilities for coding assistance in SQL.

This model enhances the context length of LLama-3 8B from 8,000 to more than 1,000,000 tokens.

Starling is a substantial language model developed through reinforcement learning based on feedback from AI, aimed at enhancing the helpfulness of chatbots.

A sophisticated language model built with 2 trillion bilingual tokens.

A version of Llama 2 that allows for a context of up to 128k tokens. This extension enhances its capacity for processing larger amounts of information.

A conversational model built on Llama 2 demonstrates strong performance across multiple benchmarks. It competes effectively with others in the field.

NVIDIA has developed a model leveraging Llama 3, which is particularly effective at conversational question answering (QA) and retrieval-augmented generation (RAG). It excels in these tasks, making it a powerful tool for generating responses and retrieving information.

BAAI's embedding model converts texts into vectors.

The sentence-transformers model is suitable for tasks such as clustering and semantic search. It can efficiently process and understand textual data for these applications.

Orca 2, developed by Microsoft Research, is an optimized version of Meta's Llama 2 models. It is specifically crafted to excel in reasoning tasks.

The BGE-M3, a recent model from BAAI, stands out for its versatility in multi-functionality, multilingual capabilities, and multi-granularity.

A general-purpose model that utilizes Llama 2.

An assistant specializing in philosophy, psychology, and personal relationships. Built on the Mistral model.

The uncensored Dolphin model, developed by Eric Hartford, is a 2.7 billion parameter model based on Microsoft's Phi language model.

A Llama 2-based model has been fine-tuned using an Orca-style dataset, originally named Free Willy.

BakLLaVA is a multimodal model that combines the Mistral 7B base model with the LLaVA architecture.

Unedited version of the Wizard LM model.

The Llama 2 model has been fine-tuned to respond to medical queries using an open-source medical dataset. This adaptation enhances its effectiveness in addressing health-related inquiries.

A version of Mistral has been expanded to accommodate context windows of 64K or 128K.

The Nous Hermes 2 model by Nous Research has now been trained on Mixtral.

An extended version of Llama 2 designed to combine general language comprehension with specialized knowledge, especially in programming and mathematics.

An efficient and cost-effective Mixture-of-Experts language model that demonstrates strength in performance.

An open-source medical large language model, tailored from Llama 2 for healthcare applications.

An excellent code generation model built on Llama2.

Nexus Raven is a 13B model specifically designed for function calling tasks. It has been fine-tuned with a focus on instruction execution.

An uncensored model based on Llama2 that offers support for a 16K context window.

A compact LLaVA model has been fine-tuned based on Phi 3 Mini.

A highly effective code instruction model has been developed by combining two existing code models.

MistralLite is a refined model derived from Mistral, designed to improve the handling of lengthy contexts. Its advanced capabilities enable better processing for extended information.

Wizard Vicuna is a 13 billion parameter model built on Llama 2, developed by MelodysDreamj.

An impressive multi-lingual general language model that competes well with Llama 3. Its performance is notably strong.

The 7B parameter text-to-SQL model was developed by MotherDuck and Numbers Station.

Falcon2 is a decoder-only model with 11 billion parameters, developed by TII and trained on 5 trillion tokens.

DBRX is a versatile, open-source large language model developed by Databricks. It is designed for a wide range of applications.

A flexible model for AI software development applications, such as code completion. It adapts to various use cases effectively.

InternLM2.5 is a 7B parameter model designed for real-world applications, showcasing exceptional reasoning abilities.