EXPLORE MODELS

DeepSeek R1

DeepSeek-R1 is an open-source first-generation reasoning model leveraging large-scale reinforcement learning to achieve state-of-the-art performance in math, code, and reasoning tasks, and includes distilled models suitable for various applications.

TRY IT

DeepSeek R1 0528

DeepSeek-R1 is an open-source first-generation reasoning model leveraging large-scale reinforcement learning to achieve state-of-the-art performance in math, code, and reasoning tasks, and includes distilled models suitable for various applications.

TRY IT

DeepSeek R1 Distill Llama 70B

Feel the power of reasoning models. This distilled model beats GPT-4o on math & matches o1-mini on coding.

TRY IT

DeepSeek V3 0324

DeepSeek-V3-0324 is an advanced language model with improved reasoning capabilities, enhanced web development support, superior Chinese writing proficiency, and refined function calling accuracy, designed to provide detailed search analysis and high-quality interactive experiences.

TRY IT

Llama 3.1 70B Instruct

The Meta Llama 3.1 collection consists of high-performing, multilingual large language models optimized for dialogue and capable of handling text and code across 8 languages, available in 8B, 70B, and 405B parameter sizes, with a focus on safety, inclusivity, and societal benefit.

TRY IT

Llama 3.1 8B Instruct

Meta Llama 3.1 is a collection of advanced, multilingual large language models designed for dialogues, available in 8B, 70B, and 405B sizes, that outperform many chat models on industry benchmarks and emphasize safe, responsible use in various applications.

TRY IT

Llama 3.2 1B Instruct

Llama 3.2 is a multilingual large language model collection from Meta, fine-tuned for dialogue and summarization tasks in multiple languages, designed for enhanced retrieval and conversational agents.

TRY IT

Llama 3.2 3B Instruct

Llama 3.2 is a multilingual large language model collection optimized for dialogue, retrieval, and summarization tasks with enhanced performance on industry benchmarks, employing supervised fine-tuning and reinforcement learning for safety and human-aligned responses.

TRY IT

Llama 3.3 70B Instruct

Meta's Llama 3.3 is a 70B parameter multilingual instruction-tuned language model designed for dialogue use, outperforming many open and closed-source models and incorporating safety features such as supervised fine-tuning and reinforcement learning with human feedback.

TRY IT

Mistral Nemo 12B Instruct

Mistral-NeMo-12B-Instruct is a 12-billion-parameter multilingual large language model designed for English-language chat applications, featuring impressive multilingual and code comprehension, with customization options via NVIDIA's NeMo Framework.

TRY IT

Osmosis Structure 0.6B

Osmosis-Structure-0.6B is a small but capable language model optimized for generating structured outputs, particularly excelling in mathematical reasoning and problem-solving tasks with impressive performance enhancements through its structured training methodology.

TRY IT

Qwen 3 30B A3B

Qwen3-30B-A3B-FP8 is a cutting-edge large language model that offers seamless switching between complex reasoning and general-purpose dialogue, excelling in reasoning, instruction-following, multilingual support, and agent integration, with a robust capacity for handling long contexts and supporting over 100 languages.

TRY IT

DeepSeek VL2 Small

DeepSeek-VL2 is an advanced series of Mixture-of-Experts Vision-Language Models designed for superior multimodal understanding tasks like visual question answering and optical character recognition, featuring three variants with varying parameter sizes.

TRY IT

DeepSeek VL2 Tiny

DeepSeek VL2 Tiny is a small-sized model that supports visual language tasks with performance comparable or even exceeding that of large-scale models.

TRY IT

Google Gemma 3

Gemma 3 is a versatile, lightweight, multimodal open-source model family by Google DeepMind, primed for text and image processing and text generation, supporting over 140 languages with a 128K context window, designed for easy deployment in resource-constrained environments.

Llama 3.2-Vision, developed by Meta, is a state-of-the-art multimodal language model optimized for image recognition, reasoning, and captioning, surpassing both open and closed models in industry benchmarks.

TRY IT