Open-Weight Models

Model

lingtai

Zesen Huang

The self-evolving Digital Scientist — a lifelong agent that grows with you and your & agent tools give you a better one-shot worker: a chat window that forgets, or a coding agent that closes with the terminal. LingTai is different — it is a Digital Scientist that lives in your project and gets better over time. It holds a question or a codebase for weeks, works with evidence and tools, records what it learns as durable knowledge and reusable skills, forms its own operating style, and delegates deep sub-problems to specialists it spawns. The work you do together becomes state the next session starts from. It is filesystem-native, not a chat window. Every agent has a home under.lingtai/; its…

Model · Image and text to text

linne-oak-1-0

Hert4

linne-oak-1.0 is a 31B decision model. Give it a state (text, a document, JSON, a conversation or an agent trajectory) and a question with a set of options, and it returns one option together with a probability for every option, in a single forward pass. It is the larger sibling of linne-fern-1.0: linne-fern is a 27B causal model that also chats, linne-oak is a dedicated decision model. - Decision Index public score 63.94, measured with the official kit (edition 0.3, all 37 public benchmarks, 140,620 requests, complete). - Reads the whole prompt at once: the full-attention layers attend in both directions over the prompt (sliding-window layers keep their native behavior), so the decision…

Model · Image classification

lite-rt_models_cadrylpro

Unicorn Who Dev

Six specialized models in 13 LiteRT files, with input/output contracts, labels, artifact hashes and CPU inference examples. The repository is public; access to its files requires manual approval on Hugging Face. Scores are specific to each model's test set and metric; they are not a common cross-task ranking. The 320 px PCB1 model is a separate compact variant of the 448 px model. See catalogue.json for exact values and source revisions. Request access on this page. After approval, authenticate and download the pinned For a binary mask or bounding boxes: Each model directory contains.tflite files, labels.txt, contract.json, format-specific metrics and CPU inference evidence. FP16 is the…

WebAssembly + WebGPU runtime binaries and the matching @litert-lm/core TypeScript SDK for running Gemma 4 (E2B, E4B, 12B, 26B-A4B, 31B) in the browser with: - Text decode with shared-KV neural Multi-Token Prediction (MTP) speculative decoding (GPUARTISAN backend, gamma 1-3, greedy token-for-token equivalence with standard decode). - Multimodal vision (SigLIP encoder on WebGPU, 70 / 140 / 280 / 560 / 1120 token budgets, zero-copy RGBA8 ImageData preprocessing, per-budget executor memory release). - Multimodal audio (Conformer encoder, streaming markers, zero-copy audio ingestion). - Streamed model loading for 2-20 GB.litertlm bundles under wasm32 (MAXIMUMMEMORY=4GB), with section filtering…

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019). - PEFT 0.10.0

This model is a PEFT/LoRA adapter single epoch fine-tuned on top of meta-llama/Llama-3.1-8B-Instruct. Finr tuning has been done on iter2 anchor-context-groundTruthjunkHandled dataset related to clinical trials. This fine tuned adapter on base LLM can be to address clinical trials related questions in general(macro/broad overview, patient eligiblity, operations etc related question). Further this can also be used for prompt engineering based on CT, or RAG generation(related to CT) based on retrieved chunks. You can load this model using bitsandbytes for 4-bit quantization and peft to apply the adapter weights. Tokenizer formatting utilizes the base model's official chat template to map…

This model is a PEFT/LoRA adapter single epoch fine-tuned on top of Fine tuning has been done on related to clinical trials. This fine tuned adapter on base LLM can be to address clinical trials related questions in general(macro/broad overview, patient eligiblity, operations etc related question). Further this can also be used for prompt engineering based on CT, or RAG generation(related to CT) based on retrieved chunks. You can load this model using bitsandbytes for 4-bit quantization and peft to apply the adapter weights. Tokenizer formatting utilizes the base model's official chat template to map query/question tokens alongside retrieved context blocks. The following configurations were…

This model is a PEFT/LoRA adapter single epoch fine-tuned on top of Fine tuning has been done on related to clinical trials. This fine tuned adapter on base LLM can be to address clinical trials related questions in general(macro/broad overview, patient eligiblity, operations etc related question). Further this can also be used for prompt engineering based on CT, or RAG generation(related to CT) based on retrieved chunks. You can load this model using bitsandbytes for 4-bit quantization and peft to apply the adapter weights. Tokenizer formatting utilizes the base model's official chat template to map query/question tokens alongside retrieved context blocks. The following configurations were…

Model

llama-guard-3-1b-llm-int8

Marcus Nogueira

Pre-quantized LLM.int8() checkpoint of meta-llama/Llama-Guard-3-1B at revision acf7aafa60f0410f8f42b1fa35e077d705892029. LLM.int8() stores the main linear weights in INT8 while dynamically processing outlier features through its higher-precision path. This is not the same scheme as static INT8 W8A8 quantization. The resolved source revision, package versions, module inventory, tensor dtypes, and file hashes are recorded in quantizationmanifest.json, runtimeenvironment.json, verification.json, and SHA256SUMS. This checkpoint passed four fixed English/Portuguese safe/unsafe smoke cases in both BF16 and FP16 with vLLM 0.31.0, bitsandbytes 0.50.2, and vllm-bnb-plugin 0.0.3 on an NVIDIA A40.…

This repository contains a patch that adds support for openPangu-2.0-Flash (92B MoE, ~6B active parameters, 512K context) to the latest llama.cpp release v0.7.0 (commit 620d2225cc436ab0b001e3200e590e3b286ef9ab, 2026-10-15). The support was ported from (which tracks an older llama.cpp base) and rebased on top of v0.7.0. - llama.cpp-openpangu-v0.7.0.diff — the full patch (43 files, +3290 / −367) - patch.sh — downloads the pinned llama.cpp commit, applies the patch and prints build/run instructions - LICENSE — GPL-3.0 (this project) with the references/licenses of the upstream projects it is built on 本 repo 是將 openPangu-2.0-Flash（92B MoE、約 6B 激活參數、512K 上下文） 的 llama.cpp 支援，從 mrexodia 的 fork…

SNI EMO random-pool seed 42. This repository contains standalone Transformers model snapshots at its root across ordered Git commits. Epochs 1–5 were committed in order. The final main revision and best tag point to the minimum validation completion CE checkpoint, epoch 2. Load the current best model with AutoModelForCausalLM.frompretrained(repoid, trustremotecode=True, torchdtype='auto', devicemap='auto'). Set revision='epoch-N' to load a specific epoch. Training sampled a pool size uniformly from 3 through 15 per example; checkpoint selection and inference use the full 15-expert pool. The model has 15 routed experts, top-3 routed selection, one always-active shared expert, and LoRA rank…

LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture. LLaVA-v1.5-7B was trained in September 2023. Paper or resources for more information: https://llava-vl.github.io/ Llama 2 is licensed under the LLAMA 2 Community License, Copyright (c) Meta Platforms, Inc. All Rights Reserved. Where to send questions or comments about the model: https://github.com/haotian-liu/LLaVA/issues The primary use of LLaVA is research on large multimodal models and chatbots. The primary intended users of the model are researchers and hobbyists in computer vision…

Model · Text generation

llm-jp-4-8b-thinking_imabari_qa_v4_reasoning_effort_grpo_v5

Ikedachin

Japanese description is available below. This model is a fully merged model created by applying GRPO training to the following supervised fine-tuned model: The immediate base model was fine-tuned on QA data containing both Imabari dialect and standard Japanese reasoning and answers. This v5 model applies additional GRPO training intended to: 1. Improve compliance with the expected reasoning-to-final-answer channel format 3. Preserve support for standard Japanese and Imabari-dialect channels 4. Preserve the low, medium, and high reasoning-effort conditions The GRPO reward used in this training run does not evaluate factual correctness, answer relevance, or dialect naturalness. It evaluates…

Model

llm-runner-aio

Nafi Ozdemir

Links Requirements Node.js and Python 3.11 must be installed. LLM-Runner-AIO.exe handles the automatic setup. After extracting the LLM-Runner-AIO.rar files, you must run run.bat first; this script installs the necessary dependencies, configures Pi Coding settings, and creates a desktop shortcut. What Does the Application Include? Searxng and llama.cpp server settings are pre-configured. You can also load functions found in the folder if you wish (e.g., EasySearch, Export to PDF/Excel/DOCX, unload llama.cpp, thinking toggle, pp/tg metrics). llama.cpp (Pre-compiled CUDA 13 + Vulkan versions) link: https://github.com/ggml-org/llama.cpp These are the versions I have configured (most models…