Open-Weight Models

This lane contains 2 distinct original checkpoint variants. Each checkpoint uses a content-addressed path and an asset manifest preserving its schema, decoder role, available format/profile binding, codec, source declarations, and companion corpus or embedding assets. Unknown bindings remain unknown. Registration and publication do not qualify numerical quality, source-only reconstruction, teacher use, proof authority, or serving. No retained checkpoint is activated for inference. The Legal 768D checkpoint is inherited initialization only; Codebase, Intent, and Security 768D lanes have no retained checkpoints. Codebase 384D envelopes retain their borrowed Security payload identity and are…

This lane contains 0 distinct original checkpoint variants. Each checkpoint uses a content-addressed path and an asset manifest preserving its schema, decoder role, available format/profile binding, codec, source declarations, and companion corpus or embedding assets. Unknown bindings remain unknown. Registration and publication do not qualify numerical quality, source-only reconstruction, teacher use, proof authority, or serving. No retained checkpoint is activated for inference. The Legal 768D checkpoint is inherited initialization only; Codebase, Intent, and Security 768D lanes have no retained checkpoints. Codebase 384D envelopes retain their borrowed Security payload identity and are…

This lane contains 2 distinct original checkpoint variants. Each checkpoint uses a content-addressed path and an asset manifest preserving its schema, decoder role, available format/profile binding, codec, source declarations, and companion corpus or embedding assets. Unknown bindings remain unknown. Registration and publication do not qualify numerical quality, source-only reconstruction, teacher use, proof authority, or serving. No retained checkpoint is activated for inference. The Legal 768D checkpoint is inherited initialization only; Codebase, Intent, and Security 768D lanes have no retained checkpoints. Codebase 384D envelopes retain their borrowed Security payload identity and are…

Model

cof-maze-checkpoints

TBD's TBD Lab

Wan2.1-T2V-1.3B 因果 AR（i2v）在迷宫导航上的训练检查点。单一仓库，按训练分目录： 即 COF-maze 的 LOGDIR 名（如 mazearscratchi2vvar4cut1preloopr2only），配置见 COF-maze 仓库 causal-forcing-changes/configs/ 或 ops/train/configs/，作业载荷见 ops/workstation/jobs/。 只上传 2000 步里程碑；从别的 run 分叉的种子检查点（软链接）不重复上传。

Model · Graph ml

cogito-estella

Jeffrey

Non-autoregressive decoder heads that turn sentence embeddings into knowledge-graph triples in one forward pass. They power the full mode of cogito-mcp, a document and code memory for agents that answers questions in a few hundred tokens. encoders.json maps the encoder to its assets and to the decode thresholds they were validated under; cogito-mcp --full reads it on first run and downloads these files from the m2m100-pool/ subfolder. The ensemble decodes 76 relations over a 20k-entity vocabulary. F1 is measured on held-out sentences whose entity and relation combinations never appeared in training, on a slice untouched by threshold selection. Each safetensors file carries its contract…

Quantized MLX weights for beshkenadze/cohere-transcribe-03-2026-mlx-fp16. - model.safetensors - config.json - tokenizer.model - tokenizerconfig.json - preprocessorconfig.json - specialtokensmap.json - keymap.json - conversionsummary.json This checkpoint has been re-validated against the current Swift and Python MLX runtimes. Verified semantic parity on an English fixture: - official CUDA reference path (transformers native Cohere ASR) Fastest and smallest, but introduces a lexical regression on the repo sample (Kaldi → Khaldi). - Generated from the Swift-compatible fp16 checkpoint beshkenadze/cohere-transcribe-03-2026-mlx-fp16. - This repository contains inference artifacts only. Refer to…

Quantized MLX weights for beshkenadze/cohere-transcribe-03-2026-mlx-fp16. - model.safetensors - config.json - tokenizer.model - tokenizerconfig.json - preprocessorconfig.json - specialtokensmap.json - keymap.json - conversionsummary.json This checkpoint has been re-validated against the current Swift and Python MLX runtimes. Verified semantic parity on an English fixture: - official CUDA reference path (transformers native Cohere ASR) Matches fp16 on the repo sample while reducing memory substantially. - Generated from the Swift-compatible fp16 checkpoint beshkenadze/cohere-transcribe-03-2026-mlx-fp16. - This repository contains inference artifacts only. Refer to the upstream Cohere model…

Model · Speech recognition

cohere-transcribe-glade

Josh Leverette

Glade is a native Swift library for on-device speech inference on Apple silicon. This repository supplies converted model assets and configuration for Glade's model-specific host runtime. Glade's runtime source is currently private. Glade conversion of Cohere Transcribe, pinned to revision 76b8b23e8607f35f0265a23d481b338fb0e26aea. The upstream model and converted weights use Apache License 2.0. Requires physical Apple silicon with macOS 27 or iOS 27 and Glade's model-specific host runtime. These graphs do not accept raw audio directly. The host provides the log-mel frontend, language prompt, token embeddings, autoregressive cache management and tokenization. See metadata.json…

Model · Text generation

coherencerl-exp7-1_5b-lora

Mark Stanley

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B. It has been trained using TRL. This model was trained with GRPO, a method introduced in DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

distilbert-base-uncased-finetuned-sst-2-english distilgpt2 Google Colab + Hugging Face

Research artifacts from an early stage project on temporally coherent colonoscopy video synthesis with exact lesion labels, evaluated by downstream polyp detection. Code and the full research record are in the companion Git repository (bundle included under code/). This is preliminary research output, not a product and not a clinical tool. Video diffusion checkpoints (57M parameter latent video DiT, mask track channel concatenated in latent space, SD f8 VAE frozen, adaLN-zero blocks, full space time attention). Each file holds model, ema, step, args. - colon-exp024c-f1-256: the main conditional model at 256px. ckpt0020000.pt is the best checkpoint (adherence Dice 0.486, judge found rate…

Model · Image to video

comfy-MiniMax-H3-workflows

Javano

Optimized ComfyUI workflows for MiniMax-H3, featuring INT8 inference, native Block Sparse Attention, Spectrum, Lightx2v LoRA, Turbo-LoRA, native Extend, Fun ControlNet Union V2V control, Face Refine, TTS, Latent Upscaler, Bridge, and Group Bypass. Comfy-Org/MiniMax-H3 https://huggingface.co/Comfy-Org/MiniMax-H3 The workflows use the native Block Sparse Attention node for optimized MiniMax-H3 attention processing. ComfyUI-Spectrum-MiniMax-H3 https://github.com/xmarre/ComfyUI-Spectrum-MiniMax-H3 Optimized sampling and MiniMax-H3 inference acceleration. lightx2v/Minimax-h3-Turbo https://huggingface.co/lightx2v/Minimax-h3-Turbo Acceleration LoRA for reduced sampling steps.…

This is a ComfyUI setup repository for RunPod. + Wan2.2 Image to Video models, text encoders, vae + Wan2.1 Image to Video 720p distilled model

Byte-identical mirrors of models hosted on individual accounts, kept so a downstream project does not break if an upstream repository is made private, renamed, or deleted. Nothing here is modified: same bytes, same checksums. Models published by institutions with a stable hosting commitment are deliberately not mirrored, since re-hosting those adds maintenance without reducing risk. The WebLLM model library is mirrored for a different reason than the rest: it is served from a GitHub raw URL on a moving branch, so the bytes behind that link can change without notice. This copy is fixed. Verify any file against its upstream with shasum -a 256.

Float16 conversions of three ONNX models for manga and comic text processing, aimed at onnxruntime-web on the WebGPU execution provider. The weights are unchanged apart from the precision cast; no retraining, pruning, or architecture edit. These exist because WebGPU rewards half precision and, in one case, punishes integer quantisation badly. Each file keeps the licence of the model it was converted from. They are distributed together as an aggregate, which GPL-3.0 section 5 permits; the GPL-3.0 files are not combined with the Apache-2.0 one. Chrome, Apple M3, onnxruntime-web on WebGPU, median of repeated runs at the input size these are run at. "Shipping" is the artifact each was converted…

static quants of https://huggingface.co/wjbmattingly/comma-lightonocr-3-4b 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…

A separate project from arghya2030/commercecore-qwen3-1.7b. No shared weights, no shared training run, no shared serving process. That release remains frozen and untouched. Status: research checkpoint, real partial progress, not a full win. Read the real results below before using this in place of a frontier model. A QLoRA rank-16 adapter on an independently-pinned original Qwen/Qwen3-1.7B base (never the Query-merged CommerceCore artifact), fine-tuned to classify the functional relationship between two product listings into substitute, complement, or unrelated. This is a dedicated, separate adapter from arghya2030/commercecore-expansion-match-v1…

A separate project from arghya2030/commercecore-qwen3-1.7b. No shared weights, no shared training run, no shared serving process. Status: research checkpoint, a rejected hypothesis, preserved as evidence. This adapter is NOT the model of record for relevance — arghya2030/commercecore-expansion-match-v1 (v1, shared adapter, 0.526 accuracy) remains the best real, adopted result. Read why this one is published anyway, below. A QLoRA rank-16 adapter on an independently-pinned original Qwen/Qwen3-1.7B base, testing the fifth of five distinct hypotheses for why the shared adapter's matchrelevance accuracy (0.526) falls short of the strongest tested frontier model, claude-sonnet-5 (0.583). Two…