Model catalog
Models their owners have shared, listed under the publisher of their weights. Each line gives what the model serves, its size, context window and license; its page gives every launch setting its owner chose and a request to call it.
25 models from 16 publishers
black-forest-labs
-
black-forest-labs/FLUX.2-dev
The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering...
- Serves
- Image generation
- Parameters
- 8.0B
- Context
- --
- License
- Apache-2.0
CohereLabs
-
CohereLabs/cohere-transcribe-03-2026
Cohere Transcribe is an open source release of a 2B parameter dedicated audio-in, text-out automatic speech recognition (ASR) model. The model supports 14 languages.
- Serves
- Chat, Transcription
- Parameters
- 2.1B
- Context
- 1K
- License
- Apache-2.0
convaiinnovations
-
convaiinnovations/laya
Python 3.10 or newer. Optional extras: laya[serve] (HTTP server), laya[mcp] (MCP server), laya[langchain] (LangChain and LangGraph), laya[onnx] (ONNX Runtime), laya[fast]...
- Serves
- System 1 decisions
- Parameters
- 743M
- Context
- 1K
- License
- Apache-2.0
-
google/gemma-4-31B-it
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and...
- Serves
- Chat, Reasoning, Multimodal input
- Parameters
- 29.7B
- Context
- 256K
- License
- Apache-2.0
gradium
-
gradium/phonon-31m-en
phonon-31m is a text-to-speech model from the Phonon family. It turns text into natural 24 kHz speech in English.
- Serves
- Speech
- Parameters
- 29M
- Context
- --
- License
- Apache-2.0
hexgrad
-
- Serves
- Speech
- Parameters
- --
- Context
- --
- License
- Apache-2.0
HiDream-ai
-
HiDream-ai/HiDream-O1-Image
HiDream-O1-Image is a natively unified image generative foundation model built on a Pixel-level Unified Transformer (UiT) without external VAEs or disjoint text encoders, which...
- Serves
- Image generation
- Parameters
- 8.2B
- Context
- --
- License
- MIT
ibm-granite
-
ibm-granite/granite-embedding-reranker-english-r2
The granite-embedding-reranker-english-r2 model uses a cross-encoder architecture to compute high-quality relevance scores between queries and documents by jointly encoding...
- Serves
- Embeddings
- Parameters
- 188M
- Context
- 8K
- License
- Apache-2.0
meta-models
-
meta-models/Muse-Glimmer-30B
Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on...
- Serves
- Chat, Reasoning, Multimodal input
- Parameters
- 28.2B
- Context
- 128K
- License
- Apache-2.0
-
meta-models/Muse-Glimmer-30B-assistant
Speculative draft for Muse-Glimmer-30B
Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on...
- Serves
- Chat, Reasoning
- Parameters
- 2.6B
- Context
- 128K
- License
- Apache-2.0
microsoft
-
microsoft/harrier-oss-v1-0.6b
harrier-oss-v1 is a family of multilingual text embedding models developed by Microsoft. The models use decoder-only architectures with last-token pooling and L2 normalization...
Popular
- Serves
- Embeddings
- Parameters
- 693M
- Context
- 32K
- License
- MIT
ornith-ai
-
ornith-ai/Ornith-1.5-35B-A3B
Chirp Chirp! 馃惁 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.
- Serves
- Chat, Reasoning
- Parameters
- 3.4B
- Context
- 256K
- License
- MIT
poolside
-
poolside/Laguna-S-2.1
Use on OpenRouter 路 Use on Vercel AI Gateway 路 Release blog post
Popular
- Serves
- Chat, Reasoning
- Parameters
- 7.3B
- Context
- 1M
- License
- openmdw-1.1
-
poolside/Laguna-XS-2.1
All benchmarking for Laguna XS 2.1 was completed using Laude Institute鈥檚 Harbor Framework with our agent harness, with a maximum of 500 steps and sandboxed execution.
- Serves
- Chat, Reasoning
- Parameters
- 2.9B
- Context
- 256K
- License
- openmdw-1.1
-
poolside/Laguna-XS-2.1-INT4
All benchmarking for Laguna XS 2.1 was completed using Laude Institute鈥檚 Harbor Framework with our agent harness, with a maximum of 500 steps and sandboxed execution.
- Serves
- Chat, Reasoning
- Parameters
- 33.3B
- Context
- 256K
- License
- openmdw-1.1
Qwen
-
Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice
Qwen3-TTS covers 10 major languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian) as well as multiple dialectal voice profiles...
- Serves
- Chat, Speech, Reasoning
- Parameters
- 2.3B
- Context
- 128K
- License
- Apache-2.0
-
Qwen/Qwen3.8-27B
Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to...
Popular
- Serves
- Chat, Reasoning, Multimodal input
- Parameters
- 23.7B
- Context
- 512K
- License
- Apache-2.0
-
Qwen/Qwen3.8-27B-FP8
Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to...
- Serves
- Chat, Reasoning, Multimodal input
- Parameters
- 23.7B
- Context
- 256K
- License
- Apache-2.0
RedHatAI
-
RedHatAI/Qwen3.5-9B-quantized.w4a16
This model is a quantized version of Qwen/Qwen3.5-9B. Evaluation results and reproduction steps are provided below.
- Serves
- Chat, Reasoning, Multimodal input
- Parameters
- 8.2B
- Context
- 256K
- License
- Apache-2.0
Tongyi-MAI
-
Tongyi-MAI/Z-Image-Turbo
鈿★笍- Image An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
- Serves
- Image generation
- Parameters
- 4.0B
- Context
- --
- License
- Apache-2.0
Independent uploads
-
This document describes the prompt encoding format used by DeepSeek-V4 series models. The encoding handles multi-turn conversations, tool calling, extended thinking...
- Serves
- Chat
- Parameters
- 83.4B
- Context
- 1M
- License
- MIT
-
Speculative draft for gemma-4-31B-it
vLLM: until Gemma4 DFlash support is merged, install vLLM from PR #41703:
- Serves
- Chat, Reasoning
- Parameters
- 4.0B
- Context
- 256K
- License
- Apache-2.0
-
Speculative draft for Laguna-XS-2.1
Use on OpenRouter 路 Use on Vercel AI Gateway 路 Release blog post
- Serves
- Chat, Reasoning
- Parameters
- 1.4B
- Context
- 1M
- License
- Not stated
-
Speculative draft for Laguna-S-2.1
DFlash speculator for poolside/Laguna-XS-2.1.
- Serves
- Chat, Reasoning
- Parameters
- 726M
- Context
- 256K
- License
- openmdw-1.1
-
Speculative draft for Qwen3.8-27B-FP8, Qwen3.8-27B
This repository contains the DFlash 2 draft model for Qwen/Qwen3.8-27B. It is not a standalone language model: it runs inside a speculative decoding server and drafts tokens...
- Serves
- Chat, Reasoning
- Parameters
- 4.2B
- Context
- 256K
- License
- Apache-2.0
No model matches this search and filter.
Reading a line
- Publisher
- The Hugging Face organization the weights come from; a model with no repository on record is listed under Independent uploads.
- Serves
- The routes the model answers, read from its capability tags. Multimodal input means it takes images, video or audio in a request.
- Context
- The window the engine launches with, in tokens. Image and audio models launch without one.
Listing a model
Upload safetensors or register a Hugging Face repository in a project, then share it. The catalog shows what the owner set and nothing they did not.