New state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.
376.2K Pulls 15 Tags Updated 2 weeks ago
QwQ is an experimental research model focused on advancing AI reasoning capabilities.
56.2K Pulls 5 Tags Updated 3 weeks ago
Llama 3.2 Vision is a collection of instruction-tuned image reasoning generative models in 11B and 90B sizes.
289.9K Pulls 9 Tags Updated 6 weeks ago
Meta's Llama 3.2 goes small with 1B and 3B models.
5M Pulls 63 Tags Updated 2 months ago
Llama 3.1 is a new state-of-the-art model from Meta available in 8B, 70B and 405B parameter sizes.
14.4M Pulls 93 Tags Updated 3 weeks ago
A high-performing open embedding model with a large token context window.
7.1M Pulls 3 Tags Updated 10 months ago
The 7B model released by Mistral AI, updated to version 0.3.
7M Pulls 84 Tags Updated 5 months ago
Meta Llama 3: The most capable openly available LLM to date
7M Pulls 68 Tags Updated 7 months ago
Gemma is a family of lightweight, state-of-the-art open models built by Google DeepMind. Updated to version 1.1
4.2M Pulls 102 Tags Updated 8 months ago
Qwen 1.5 is a series of large language models by Alibaba Cloud spanning from 0.5B to 110B parameters
4.2M Pulls 379 Tags Updated 7 months ago
Qwen2 is a new series of large language models from Alibaba group
4M Pulls 97 Tags Updated 3 months ago
Phi-3 is a family of lightweight 3B (Mini) and 14B (Medium) state-of-the-art open models by Microsoft.
2.8M Pulls 72 Tags Updated 4 months ago
Llama 2 is a collection of foundation language models ranging from 7B to 70B parameters.
2.7M Pulls 102 Tags Updated 12 months ago
Qwen2.5 models are pretrained on Alibaba's latest large-scale dataset, encompassing up to 18 trillion tokens. The model supports up to 128K tokens and has multilingual support.
2.5M Pulls 133 Tags Updated 3 months ago
🌋 LLaVA is a novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding. Updated to version 1.6.
2.2M Pulls 98 Tags Updated 10 months ago
Google Gemma 2 is a high-performing and efficient model available in three sizes: 2B, 9B, and 27B.
2.2M Pulls 94 Tags Updated 4 months ago
A large language model that can use text prompts to generate and discuss code.
1.6M Pulls 199 Tags Updated 5 months ago
The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing.
951.6K Pulls 196 Tags Updated 5 weeks ago
The TinyLlama project is an open endeavor to train a compact 1.1B Llama model on 3 trillion tokens.
754.9K Pulls 36 Tags Updated 11 months ago
A state-of-the-art 12B model with 128k context length, built by Mistral AI in collaboration with NVIDIA.
752.7K Pulls 17 Tags Updated 4 months ago
State-of-the-art large embedding model from mixedbread.ai
706.3K Pulls 4 Tags Updated 7 months ago
StarCoder2 is the next generation of transparently trained open code LLMs that comes in three sizes: 3B, 7B and 15B parameters.
542.9K Pulls 67 Tags Updated 3 months ago
A set of Mixture of Experts (MoE) model with open weights by Mistral AI in 8x7b and 8x22b parameter sizes.
503.7K Pulls 70 Tags Updated yesterday
Uncensored, 8x7b and 8x22b fine-tuned models based on the Mixtral mixture of experts models that excels at coding tasks. Created by Eric Hartford.
456.2K Pulls 70 Tags Updated 2 days ago
CodeGemma is a collection of powerful, lightweight models that can perform a variety of coding tasks like fill-in-the-middle code completion, code generation, natural language understanding, mathematical reasoning, and instruction following.
445.8K Pulls 85 Tags Updated 5 months ago
An open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks.
434.4K Pulls 64 Tags Updated 3 months ago
Phi-2: a 2.7B language model by Microsoft Research that demonstrates outstanding reasoning and language understanding capabilities.
427.3K Pulls 18 Tags Updated 11 months ago
Uncensored Llama 2 model by George Sung and Jarrad Hope.
388.3K Pulls 34 Tags Updated 14 months ago
DeepSeek Coder is a capable coding model trained on two trillion code and natural language tokens.
386.3K Pulls 102 Tags Updated 12 months ago
A suite of text embedding models by Snowflake, optimized for performance.
342.8K Pulls 16 Tags Updated 8 months ago
State of the art large language model from Microsoft AI with improved performance on complex chat, multilingual, reasoning and agent use cases.
304.4K Pulls 22 Tags Updated 8 months ago
The uncensored Dolphin model based on Mistral that excels at coding tasks. Updated to version 2.8.
290.4K Pulls 120 Tags Updated 8 months ago
Dolphin 2.9 is a new model with 8B and 70B sizes by Eric Hartford based on Llama 3 that has a variety of instruction, conversational, and coding skills.
256.2K Pulls 53 Tags Updated 7 months ago
Yi 1.5 is a high-performing, bilingual language model.
252.3K Pulls 174 Tags Updated 7 months ago
Command R is a Large Language Model optimized for conversational interaction and long context tasks.
250.6K Pulls 32 Tags Updated 3 months ago
A general-purpose model ranging from 3 billion parameters to 70 billion, suitable for entry-level hardware.
249.5K Pulls 119 Tags Updated 14 months ago
A LLaVA model fine-tuned from Llama 3 Instruct with better scores in several benchmarks.
233.2K Pulls 4 Tags Updated 7 months ago
Zephyr is a series of fine-tuned versions of the Mistral and Mixtral models that are trained to act as helpful assistants.
228K Pulls 40 Tags Updated 8 months ago
A lightweight AI model with 3.8 billion parameters with performance overtaking similarly and larger sized models.
199.6K Pulls 17 Tags Updated 3 months ago
Embedding models on very large sentence level datasets.
184.5K Pulls 10 Tags Updated 7 months ago
Codestral is Mistral AI’s first-ever code model designed for code generation tasks.
180.6K Pulls 17 Tags Updated 3 months ago
StarCoder is a code generation model trained on 80+ programming languages.
174.2K Pulls 100 Tags Updated 14 months ago
General use chat model based on Llama and Llama 2 with 2K to 16K context sizes.
166.7K Pulls 111 Tags Updated 14 months ago
A family of open foundation models by IBM for Code Intelligence
165.5K Pulls 162 Tags Updated 3 months ago
Mistral OpenOrca is a 7 billion parameter model, fine-tuned on top of the Mistral 7B model using the OpenOrca dataset.
162.3K Pulls 17 Tags Updated 14 months ago
🪐 A family of small models with 135M, 360M, and 1.7B parameters, trained on a new high-quality dataset.
156.6K Pulls 94 Tags Updated 4 months ago
Wizard Vicuna Uncensored is a 7B, 13B, and 30B parameter model based on Llama 2 uncensored by Eric Hartford.
151.7K Pulls 49 Tags Updated 14 months ago
BGE-M3 is a new model from BAAI distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity.
146.2K Pulls 3 Tags Updated 4 months ago
Llama 2 based model fine tuned to improve Chinese dialogue ability.
141.6K Pulls 35 Tags Updated 14 months ago
A versatile model for AI software development scenarios, including code completion.
128.6K Pulls 17 Tags Updated 5 months ago
A family of open-source models trained on a wide variety of data, surpassing ChatGPT on various benchmarks. Updated to version 3.5-0106.
123.5K Pulls 50 Tags Updated 11 months ago
Aya 23, released by Cohere, is a new family of state-of-the-art, multilingual models that support 23 languages.
122.4K Pulls 33 Tags Updated 7 months ago
CodeQwen1.5 is a large language model pretrained on a large amount of code data.
120.8K Pulls 30 Tags Updated 6 months ago
The powerful family of models by Nous Research that excels at scientific discussion and coding tasks.
117.7K Pulls 33 Tags Updated 11 months ago
Command R+ is a powerful, scalable large language model purpose-built to excel at real-world enterprise use cases.
113.8K Pulls 21 Tags Updated 3 months ago
Mistral Large 2 is Mistral's new flagship model that is significantly more capable in code generation, mathematics, and reasoning with 128k context window and support for dozens of languages.
110.8K Pulls 32 Tags Updated 4 weeks ago
Stable Code 3B is a coding model with instruct and code completion variants on par with models such as Code Llama 7B that are 2.5x larger.
110.6K Pulls 36 Tags Updated 9 months ago
State-of-the-art code generation model
110.4K Pulls 67 Tags Updated 11 months ago
An experimental 1.1B parameter model trained on the new Dolphin 2.8 dataset by Eric Hartford and based on TinyLlama.
110.1K Pulls 18 Tags Updated 10 months ago
OpenHermes 2.5 is a 7B model fine-tuned by Teknium on Mistral with fully open datasets.
108.2K Pulls 35 Tags Updated 11 months ago
Qwen2 Math is a series of specialized math language models built upon the Qwen2 LLMs, which significantly outperforms the mathematical capabilities of open-source models and even closed-source models (e.g., GPT4o).
107.5K Pulls 52 Tags Updated 3 months ago
A strong multi-lingual general language model with competitive performance to Llama 3.
105.8K Pulls 32 Tags Updated 5 months ago
Stable LM 2 is a state-of-the-art 1.6B and 12B parameter language model trained on multilingual data in English, Spanish, German, Italian, French, Portuguese, and Dutch.
101.1K Pulls 84 Tags Updated 7 months ago
BakLLaVA is a multimodal model consisting of the Mistral 7B base model augmented with the LLaVA architecture.
100.5K Pulls 17 Tags Updated 12 months ago
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.
98.8K Pulls 17 Tags Updated 3 months ago
An advanced language model crafted with 2 trillion bilingual tokens.
93.9K Pulls 64 Tags Updated 12 months ago
This model extends LLama-3 8B's context length from 8k to over 1m tokens.
93.1K Pulls 35 Tags Updated 7 months ago
Model focused on math and logic problems
93K Pulls 64 Tags Updated 12 months ago
moondream2 is a small vision language model designed to run efficiently on edge devices.
87.9K Pulls 18 Tags Updated 7 months ago
A fine-tuned model based on Mistral with good coverage of domain and language.
86.4K Pulls 50 Tags Updated 12 months ago
A model from NVIDIA based on Llama 3 that excels at conversational question answering (QA) and retrieval-augmented generation (RAG).
81.9K Pulls 35 Tags Updated 7 months ago
SQLCoder is a code completion model fined-tuned on StarCoder for SQL generation tasks
81.4K Pulls 48 Tags Updated 10 months ago
Conversational model based on Llama 2 that performs competitively on various benchmarks.
81.2K Pulls 80 Tags Updated 13 months ago
General use models based on Llama and Llama 2 from Nous Research.
78.8K Pulls 63 Tags Updated 13 months ago
Code generation model based on Code Llama.
77.6K Pulls 49 Tags Updated 12 months ago
An extension of Llama 2 that supports a context of up to 128k tokens.
75.7K Pulls 67 Tags Updated 13 months ago
A 7B and 15B uncensored variant of the Dolphin model family that excels at coding, based on StarCoder2.
75.6K Pulls 35 Tags Updated 8 months ago
General use model based on Llama 2.
74.1K Pulls 73 Tags Updated 13 months ago
A strong, economical, and efficient Mixture-of-Experts language model.
72.7K Pulls 34 Tags Updated 6 months ago
Hermes 3 is the latest version of the flagship Hermes series of LLMs by Nous Research
64.9K Pulls 65 Tags Updated 5 days ago
Starling is a large language model trained by reinforcement learning from AI feedback focused on improving chatbot helpfulness.
64.7K Pulls 36 Tags Updated 8 months ago
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.
64.4K Pulls 67 Tags Updated 3 months ago
A companion assistant trained in philosophy, psychology, and personal relationships. Based on Mistral.
64.4K Pulls 49 Tags Updated 14 months ago
InternLM2.5 is a 7B parameter model tailored for practical scenarios with outstanding reasoning capability.
64K Pulls 65 Tags Updated 4 months ago
A large language model built by the Technology Innovation Institute (TII) for use in summarization, text generation, and chat bots.
63.8K Pulls 38 Tags Updated 14 months ago
A compact, yet powerful 10.7B large language model designed for single-turn conversation.
63.7K Pulls 32 Tags Updated 12 months ago
SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters.
60.9K Pulls 49 Tags Updated 7 weeks ago
A new small LLaVA model fine-tuned from Phi 3 Mini.
60.1K Pulls 4 Tags Updated 7 months ago
Athene-V2 is a 72B parameter model which excels at code completion, mathematics, and log extraction tasks.
59.3K Pulls 17 Tags Updated 5 weeks ago
Orca 2 is built by Microsoft research, and are a fine-tuned version of Meta's Llama 2 models. The model is designed to excel particularly in reasoning.
59.1K Pulls 33 Tags Updated 13 months ago
A series of multimodal LLMs (MLLMs) designed for vision-language understanding.
58K Pulls 17 Tags Updated 4 weeks ago
Llama 2 based model fine tuned on an Orca-style dataset. Originally called Free Willy.
56.7K Pulls 49 Tags Updated 13 months ago
Mistral Small is a lightweight model designed for cost-effective use in tasks like translation and summarization.
53.9K Pulls 17 Tags Updated 3 months ago
2.7B uncensored Dolphin model by Eric Hartford, based on the Phi language model by Microsoft Research.
53K Pulls 15 Tags Updated 12 months ago
Uncensored version of Wizard LM model
49K Pulls 18 Tags Updated 14 months ago
A commercial-friendly small language model by NVIDIA optimized for roleplay, RAG QA, and function calling.
45.5K Pulls 17 Tags Updated 3 months ago
An extension of Mistral to support context windows of 64K or 128K.
43.5K Pulls 33 Tags Updated 13 months ago
An expansion of Llama 2 that specializes in integrating both general language understanding and domain-specific knowledge, particularly in programming and mathematics.
43.3K Pulls 33 Tags Updated 11 months ago
Fine-tuned Llama 2 model to answer medical questions based on an open source medical dataset.
42K Pulls 17 Tags Updated 13 months ago
Open-source medical large language model adapted from Llama 2 to the medical domain.
41.2K Pulls 22 Tags Updated 12 months ago
A series of models from Groq that represent a significant advancement in open-source AI capabilities for tool use/function calling.
41.1K Pulls 33 Tags Updated 5 months ago
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.
41K Pulls 17 Tags Updated 2 months ago
Nexus Raven is a 13B instruction tuned model for function calling tasks.
39.6K Pulls 32 Tags Updated 11 months ago
The Nous Hermes 2 model from Nous Research, now trained over Mixtral.
36.2K Pulls 18 Tags Updated 2 days ago
Great code generation model based on Llama2.
34.9K Pulls 19 Tags Updated 13 months ago
Uncensored Llama2 based model with support for a 16K context window.
33.5K Pulls 18 Tags Updated 12 months ago
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.
31.9K Pulls 33 Tags Updated 4 weeks ago
🎩 Magicoder is a family of 7B parameter models trained on 75K synthetic instruction data using OSS-Instruct, a novel approach to enlightening LLMs with open-source code snippets.
30.1K Pulls 18 Tags Updated 12 months ago
A lightweight chat model allowing accurate, and responsive output without requiring high-end hardware.
29.4K Pulls 17 Tags Updated 12 months ago
A high-performing code instruct model created by merging two existing code models.
29.1K Pulls 16 Tags Updated 13 months ago
Falcon2 is an 11B parameters causal decoder-only model built by TII and trained over 5T tokens.
28.8K Pulls 17 Tags Updated 7 months ago
Wizard Vicuna is a 13B parameter model based on Llama 2 trained by MelodysDreamj.
27.9K Pulls 17 Tags Updated 14 months ago
MistralLite is a fine-tuned model based on Mistral with enhanced capabilities of processing long contexts.
27.5K Pulls 17 Tags Updated 13 months ago
MathΣtral: a 7B model designed for math reasoning and scientific discovery by Mistral AI.
26.9K Pulls 17 Tags Updated 5 months ago
7B parameter text-to-SQL model made by MotherDuck and Numbers Station.
26.7K Pulls 17 Tags Updated 10 months ago
MegaDolphin-2.2-120b is a transformation of Dolphin-2.2-70b created by interleaving the model with itself.
24.1K Pulls 19 Tags Updated 11 months ago
Solar Pro Preview: an advanced large language model (LLM) with 22 billion parameters designed to fit into a single GPU
23.8K Pulls 18 Tags Updated 3 months ago
A series of models that convert HTML content to Markdown content, which is useful for content conversion tasks.
23.7K Pulls 33 Tags Updated 3 months ago
A top-performing mixture of experts model, fine-tuned with high-quality data.
23.1K Pulls 18 Tags Updated 11 months ago
A 7B chat model fine-tuned with high-quality data and based on Zephyr.
22.4K Pulls 18 Tags Updated 11 months ago
Merge of the Open Orca OpenChat model and the Garage-bAInd Platypus 2 model. Designed for chat and code generation.
22.4K Pulls 17 Tags Updated 13 months ago
A language model created by combining two fine-tuned Llama 2 70B models into one.
22K Pulls 16 Tags Updated 13 months ago
The IBM Granite 1B and 3B models are the first mixture of experts (MoE) Granite models from IBM designed for low latency usage.
20.8K Pulls 33 Tags Updated 4 weeks ago
An open large reasoning model for real-world solutions by the Alibaba International Digital Commerce Group (AIDC-AI).
20.6K Pulls 5 Tags Updated 2 weeks ago
Cohere For AI's language models trained to perform well across 23 different languages.
18.8K Pulls 33 Tags Updated 8 weeks ago
A 3.8B model fine-tuned on a private high-quality synthetic dataset for information extraction, based on Phi-3.
18.4K Pulls 17 Tags Updated 5 months ago
DBRX is an open, general-purpose LLM created by Databricks.
17K Pulls 7 Tags Updated 8 months ago
Embedding model from BAAI mapping texts to vectors.
16.6K Pulls 3 Tags Updated 4 months ago
An open weights function calling model based on Llama 3, competitive with GPT-4o function calling capabilities.
15.7K Pulls 17 Tags Updated 5 months ago
A robust conversational model designed to be used for both chat and instruct use cases.
14K Pulls 7 Tags Updated 13 months ago
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.
12.8K Pulls 49 Tags Updated 2 months ago
An upgraded version of DeekSeek-V2 that integrates the general and coding abilities of both DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct.
12.8K Pulls 7 Tags Updated 3 months ago
A state-of-the-art fact-checking model developed by Bespoke Labs.
12.6K Pulls 17 Tags Updated 3 months ago
Sentence-transformers model that can be used for tasks like clustering or semantic search.
12.4K Pulls 3 Tags Updated 4 months ago
Llama Guard 3 is a series of models fine-tuned for content safety classification of LLM inputs and responses.
12.4K Pulls 33 Tags Updated 2 months ago
OpenCoder is an open and reproducible code LLM family which includes 1.5B and 8B models, supporting chat in English and Chinese languages.
11.5K Pulls 9 Tags Updated 4 weeks ago
Tülu 3 is a leading instruction following model family, offering fully open-source data, code, and recipes by the The Allen Institute for AI.
6,769 Pulls 9 Tags Updated yesterday
Snowflake's frontier embedding model. Arctic Embed 2.0 adds multilingual support without sacrificing English performance or scalability.
5,161 Pulls 3 Tags Updated 2 weeks ago
A family of efficient AI models under 10B parameters performant in science, math, and coding through innovative training techniques.
4,234 Pulls 17 Tags Updated 4 days ago
EXAONE 3.5 is a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.
3,806 Pulls 13 Tags Updated 11 days ago
The IBM Granite 2B and 8B models are text-only dense LLMs trained on over 12 trillion tokens of data, demonstrated significant improvements over their predecessors in performance and speed in IBM’s initial testing.
3,418 Pulls 33 Tags Updated 3 days ago
The IBM Granite Guardian 3.0 2B and 8B models are designed to detect risks in prompts and/or responses.
3,090 Pulls 10 Tags Updated 4 weeks ago
The IBM Granite 1B and 3B models are long-context mixture of experts (MoE) Granite models from IBM designed for low latency usage.
1,805 Pulls 33 Tags Updated 4 days ago
Sailor2 are multilingual language models made for South-East Asia. Available in 1B, 8B, and 20B parameter sizes.
1,527 Pulls 13 Tags Updated 2 weeks ago
The IBM Granite Embedding 30M and 278M models models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases.
951 Pulls 6 Tags Updated 4 days ago