Open-Weight Models

Model · Text generation

Frontis-MA1-30B-SFT

Frontis AI

This is the supervised fine-tuning checkpoint of the Frontis-MA1 family, before OpenMLE reinforcement learning. It is intended for research on machine learning engineering and program transformation. The actual 24,350-row training input is distinct from the 26,259-example public SFT trace collection. Dataset cutoff configuration was 32,768 tokens; this does not establish validated long-context inference performance. No standalone SFT benchmark claim is made by this release. The separately released RL model is FrontisAI/Frontis-MA1-30B. Pin the published commit SHA for reproducibility. This model requires substantial memory; use the dependency versions recorded in the release validation…

Model · Image and text to text

Frontis-MA1-35B-SFT

Frontis AI

This is the supervised fine-tuning checkpoint of the Frontis-MA1 family, before OpenMLE reinforcement learning. It is intended for research on machine learning engineering and program transformation. The actual 24,350-row training input is distinct from the 26,259-example public SFT trace collection. Dataset cutoff configuration was 32,768 tokens; this does not establish validated long-context inference performance. No standalone SFT benchmark claim is made by this release. The separately released RL model is FrontisAI/Frontis-MA1-35B. Pin the published commit SHA for reproducibility. This model requires substantial memory; use the dependency versions recorded in the release validation…

A fine-tuned google/functiongemma-270m-it model for naming chat sessions from a user's opening message. It produces a setsessiontitle function call containing a concise title, rather than answering the message. The saved model is in functiongemmatitlemodel/, including full model weights, tokenizer, generation configuration, and chat template. - Generate a title of 3–7 words in sentence case. - Describe the user's task or conversation topic. - Preserve technical terms. - Return only a setsessiontitle tool call. Example output from the saved tuned evaluation: This is FunctionGemma's tool-call syntax, not JSON. The application must parse and validate the call before using the title. Install…

Model · Unconditional image generation

FuseReg

Hongyang Du

Official checkpoints for FuseReg. FuseReg replaces heuristic encoder-layer fusion in representation autoencoders (RAEs) with training over random subsets of encoder layers, so a single pixel decoder reconstructs reliably under any layer-subset fusion and pairs better with the diffusion generator. The repository currently contains 16 inference checkpoints: eight DINOv3-L k=23 decoders, two DINOv3-L k=7 decoders, and six DiT-XL generators. The k=7 p=0.3, four EUPE, and four SigLIP variants have not been published. A 25-file candidate collection is tracked in the checkpoint manifest; only entries in its files array are available for download. Checkpoint manifest records file size, SHA-256…

static quants of https://huggingface.co/endless-frontier/Fx-Work For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant…

weighted/imatrix quants of https://huggingface.co/Hastagaras/G4-Jamet-26B-A4B-MK-1 For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/G4-Jamet-26B-A4B-MK-1-GGUF This is a vision model - mmproj files (if any) will be in the static repository. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on…

weighted/imatrix quants of https://huggingface.co/vulture3/G4-WanaMeroQueen-v3-31B-heretic For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/G4-WanaMeroQueen-v3-31B-heretic-GGUF This is a vision model - mmproj files (if any) will be in the static repository. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are…

Model · Image to video

GAE-D64-1B

ARC Lab, Tencent

(hiddensize=[768, 2048], depth=[28, 6]) over a 64-channel geometry-native latent, at 672×378 and V=81, with metric Plücker camera conditioning and Qwen3-0.6B text. This repo is self-contained for GAE-64 inference: codec, per-channel latent statistics, DA3 feature-level stats, and the flow transformer. The frozen DA3-GIANT encoder / DPT head comes from Or, after the files are in ckpts/: Image-to-video uses 50 Euler steps and CFG 2 with one reference view. - codec/ — safetensors + config.json - gae64.pt — drop-in checkpoint for gae.loadcodec - latentstatsgae64.pt — per-channel latent mean/std - da3statsgiant5ds.tar — DA3 feature-level normalization stats - transformer/ — flow safetensors +…

Weights for the three-stage GATOR cascade, which reconstructs a textured mesh from several masked RGB views of one object. Each.statedict.ckpt holds only the fp32 model statedict of that training step, without optimizer or EMA state. Each config.yaml is the training configuration of its run; its dataset paths are not used at inference. t5gemma-2-1b-1b-text-only-dd0a268/ holds the text encoder of google/t5gemma-2-1b-1b at revision dd0a2683227859151b1730ca3a63087df5b5f39b: the model.encoder. weights in bfloat16 plus the tokenizer. taoyuantextencodermanifest.json records the export. Gemma is provided under and subject to the Gemma Terms of Use found at ai.google.dev/gemma/terms; see the…

Model

GDN_pico_ByT5_tokenizer

Kurogane Himaki

use attahed notebook.

Model

GLM-4.7-Flash-Hauhau-Aggressive-W6A8

Cassidy Barton

Native compressed-tensors W6A8 checkpoint published by Xananthium. Credit and upstream terms: HauhauCS/GLM-4.7-Flash-Uncensored-HauhauCS-Aggressive. The source model card is preserved in READMEORIGINAL.md. Quantization provenance and retained high-precision components are recorded in conversion-receipt.json. I wanted W6A8, so that is what we built: six-bit stored weights, dynamic eight-bit activations, and vLLM doing the serving. Humming packs the INT6 values across INT32 word boundaries and unpacks them for the 3090's supported INT8 matrix math. The cards do not need a native six-bit Tensor Core instruction. The extra unpacking and scale work is part of the cost, and the speed tests…

GLM-5.3 uses the same base model as GLM-5.2 — every gain comes from post-training. Compared with GLM-5.2, it is much better at complex coding and long-horizon tasks: GLM-5.3 supports deployment with the following frameworks. Feel free to try them out: - SGLang — see cookbook - vLLM — see recipes - TokenSpeed — see here - Transformers — see transformers docs - KTransformers — see tutorial - Unsloth — see guide - For deployment on the Ascend NPU platform, inference frameworks such as vLLM-Ascend, xLLM and SGLang are supported — see here. - GLM-5.3 supports controlling the thinking budget through the reasoningeffort parameter, which accepts three levels: low, high, and max. It defaults to max…

GLM-5.3 uses the same base model as GLM-5.2 — every gain comes from post-training. Compared with GLM-5.2, it is much better at complex coding and long-horizon tasks: GLM-5.3 supports deployment with the following frameworks. Feel free to try them out: - SGLang — see cookbook - vLLM — see recipes - TokenSpeed — see here - Transformers — see transformers docs - KTransformers — see tutorial - Unsloth — see guide - For deployment on the Ascend NPU platform, inference frameworks such as vLLM-Ascend, xLLM and SGLang are supported — see here. - GLM-5.3 supports controlling the thinking budget through the reasoningeffort parameter, which accepts three levels: low, high, and max. It defaults to max…

Model · Text generation

GLM-5.3-DSpark

Alaya NeW

This repository releases a DSpark draft model for GLM-5.3. It is a speculator, not a standalone LLM: pair it with a GLM-5.3 target (BF16 or FP8) in SGLang or vLLM. Requires a recent SGLang or vLLM build with native DSpark support. This draft is for GLM-5.3 only. --speculative-num-draft-tokens is block size + 1. The draft is bf16 dense. Keep the draft KV cache dtype on auto so it does not inherit the target's FP8 MLA KV cache. - Context-length numbers are the mean of 3 runs (50 prompts, SPEED-Bench low-entropy throughput) - Qualitative numbers are one run on SPEED-Bench categories (80 prompts each) Acceptance length is the per-request mean of completion tokens divided by verification steps.…

Model

GLM-5.3-Flash-Distill-Qwen-7B-LoRA

Sscl

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.