Search the full model catalog.

At 1.54B parameters, the smallest Qwen2.5 chat model to pass 50 on both MATH and HumanEval in Qwen's tables (55.2 and 61.6). It is licensed under Apache 2.0.

Describes images and answers questions about them in plain text. It is the image checkpoint of DAMO-NLP-SG's VideoLLaMA 3, with a SigLIP-derived encoder that takes images of varying sizes.

Answers questions about a video clip or describes it in text. DAMO-NLP-SG built it on Qwen2.5-7B with a SigLIP-based vision encoder, and its training centers on image-text data.

Reads Turkish license plates in a photo and returns them as text, along with a copy of the image where each car is numbered and each plate is boxed.

Edits an existing image from a written instruction such as "add a beautiful sunset", with no mask to draw. Fine-tuned from Stable Diffusion 1.5 by Tim Brooks, Aleksander Holynski and Alexei A. Efros.

PixArt-Σ consists of pure transformer blocks for latent diffusion: It can directly generate 1024px, 2K and 4K images from text prompts within a single sampling process.

Generate synchronized audio for videos using AI. Enhance video content with matching sound effects or create custom audio from text descriptions.

Answers questions from files you upload, such as PDF, Word, PowerPoint or CSV. The chat model you choose writes each answer from the passages that best match the question.

Finds printed text in an image and returns it as plain text, one line per detected region. Runs the open-source EasyOCR library with 17 selectable languages.

Qwen2's 7B chat model from June 2024, aligned with supervised fine-tuning and DPO. Qwen reports 79.9 on HumanEval and 77.2 on C-Eval; on Wiro it takes a prompt plus session history.

Writes English and Chinese math solutions either as plain derivations or with Python-assisted steps. Qwen reports 83.6 on MATH for the 7B model with chain-of-thought, and 85.3 when an interpreter runs its Python.

A 1.5B-parameter Qwen2.5-Math chat model that solves English and Chinese math problems step by step and boxes the final answer. Qwen reports 79.7 on MATH for it when an interpreter runs the Python it writes.

A 7.61B-parameter Qwen2.5-Coder chat model for generating and fixing code. On Wiro it reads up to 32,768 input tokens per request, close to twice the limit of its 32B sibling.

The 0.5B-parameter chat model of Qwen1.5, the smallest size in the series, post-trained with supervised fine-tuning and DPO. It suits short replies and rewrites where the prompt supplies the facts.

Microsoft's 14B-parameter open model for English math and reasoning, built for memory- or compute-limited setups and trained with synthetic textbook-style data. MIT licensed.

The 32.5B-parameter flagship of Qwen2.5-Coder, instruction-tuned for writing and repairing code. Qwen reports 73.7 on the Aider code-repair benchmark and 65.9 on McEval, which spans more than 40 languages.

Gemma 2 2B IT is the smallest instruction-tuned Gemma 2 model from Google, with about 2.6B parameters, for short English replies and rewrites where the prompt supplies the facts.

Gemma 2 9B IT is Google's 9-billion-parameter instruction-tuned Gemma 2 model for English chat and writing. Google reports its pretrained base at 68.6 on GSM8K math, against 23.9 for the 2B.

Meta's 8B-parameter Llama 3 model tuned for English assistant chat. It answers with text and code inside an 8k-token context and can keep a separate chat history per user and session.

Mistral AI's 7B chat model, fine-tuned for instructions from Mistral-7B-v0.3. This version added function calling and moved to the v3 tokenizer with a 32,768-token vocabulary.