Open-Weight Models

Model · Reinforcement learning

energydecision-dt-v2-sdp

Victor U

The shipped / preferred model of the energydecision benchmark. A standalone Decision Transformer distilled from an honest (non-clairvoyant), degradation-aware SDP-planning teacher — no solver in the control loop (the Jt(soc) prompt is precomputed) — that beats the PPO reference on all four identity surfaces and remains strong under the market-impact benchmark. (Removed 2026-09: the behaviour-cloning-era files aemodtfcasmodel.pt and aemodtfcasbestcheckpoint.pt.) All four identity surfaces beat PPO with paired-difference 95% CIs excluding zero. Methodology, the physics fixes, and the full experiment ladder: report.md §8.2.10/§8.2.11 and docs/knownissues.md. The model card's numbers above are…

Model

energydecision-waypoint-dt

Victor U

A small Decision Transformer that predicts target-SOC waypoints for the hierarchical planning teacher used to build the corpus. The teacher is a two-stage hierarchy: 1. This model predicts a coarse target-SOC waypoint trajectory (K = 8 waypoints) over the episode. 2. The honest SDP executor (src/aemosdpexecutor.py) plans energy dispatch between waypoints under a seasonal price forecast. 3. A greedy FCAS bidder allocates residual headroom to the highest-priced raise/lower service each step. The resulting (observation, action, reward) triples form the SDP-teacher corpus. targets produced from perfect-foresight Oracle LP paths. - Label generator (GitHub, not shipped here): The label dataset is…

This model is a fine-tuned version of fpadovani/eng-latn-10mb-ppt-Dp-100mb-packedseed3407. It has been trained using TRL. This model was trained with SFT.

This model is a fine-tuned version of fpadovani/eng-latn-10mb-ppt-Dp-10mb-packedseed3407. It has been trained using TRL. This model was trained with SFT.

An English human/AI detector for local Python, CLI and browser use, with calibrated probabilities, explicit abstention and exploratory source rankings. AI includes direct generation and substantive AI rewriting, including a human original rewritten by DIPPER. Results are not an AI-written word percentage, proof of authorship or authenticated model identity. The frozen binary model passed all preregistered point criteria on four necessary final cohorts and both required rewrite-origin views. This is a measured-scope result, not universal accuracy or a population/per-domain 1% false-positive guarantee. Family and nominal-version assertions are disabled; both acceptedfamily and acceptedmodel…

Model · Speech recognition

ep-hindi-asr

Epoch Discovery

ep-hindi-asr is a Hindi speech recognition model built on the audio encoder of EmbeddingGemma 2. A CTC head reads a learned mix of the encoder's layers and predicts Gemma tokenizer units. Only the top two encoder layers, the layer mix and the head are trained; the rest of the encoder is used as released by Google DeepMind. The model was trained in two stages on a single NVIDIA A10: 300 hours of IndicVoices Hindi, then continued fine-tuning on IndicVoices, Vaani, Kathbath and Gram Vaani with noise augmentation. It decodes in one pass with no autoregressive decoder, and transcribes about 800 times faster than real time with batched input. All results use the released checkpoint (stage 2, step…

Trained with LeRobot.

Trained with LeRobot.

Model

ermine-megatron

Chen-Hao Chao

Megatron-LM checkpoints for Masked Diffusion Models (MDM) with Mixture-of-Experts (MoE) at various scales. Each.tar file contains a torchdist Megatron-LM distributed checkpoint.

Model

ermine-olmoe

Chen-Hao Chao

This repository hosts intermediate training checkpoints (step 170,000) for two ~1B-active-parameter masked diffusion language models (MDMs) built on an OLMoE-style Mixture-of-Experts transformer backbone. The two checkpoints share the same architecture and data but differ in the schedule used to sample the diffusion timestep t during training. Code: https://github.com/chen-hao-chao/olmoe-mdm (fork of AI2's OLMo) Monte-Carlo ELBO objective rather than next-token autoregression. The model has bidirectional attention (bidirectional: true) and is trained to denoise randomly masked token subsets. position embeddings, RMSNorm, and SwiGLU activations. - dmodel: 2240 - nlayers: 18 - nheads: 35…

Model

escha-runtime-qwen3dense

Escha Labs

The serving runtime for Escha 2-/3-bit (escha) quantized models of the qwen35 dense architecture (Qwen3.8-27B and siblings). One repo per model architecture, one directory per engine — this architecture currently has one engine, sglang/. The engine is a fork of SGLang bundled inside the wheel, running the Escha CUDA kernels. No separate sglang install is needed, and none should be present — the wheel ships its own. Full detail, including the per-GPU cookbook and troubleshooting: Then check the stack and the endpoint: This is a reasoning model. With thinking on, the reasoning arrives in reasoningcontent and the answer in content — read both, or you will see half the response. Two per-request…

Model

escha-runtime-qwen3moe

Escha Labs

Serving runtimes for Escha 2-/3-bit (eschamoe) quantized models of the qwen35moe architecture (Qwen3.5 / Qwen3.6 Mixture-of-Experts, 256 experts). One repo per model architecture, one directory per engine — pick the engine that fits your workload: Neither engine is uniformly faster. ZML wins sustained decode on long single-stream answers and is far simpler to deploy; SGLang wins short replies, wins time-to-first-token on long prompts, and is the only option for concurrency, tool calling or structured output. If unsure: multi-user, agents-at-scale, or structured output → SGLang; long-form single-user generation on your own GPU → ZML. The ZML engine is built on ZML (Zig + MLIR/XLA); the…

Model · Text generation

esp32-s3-tinystories

Nicholas Wilde

Quantized INT4 weights designed to run locally on ESP32-S3 with 16MB Flash and Octal PSRAM. This model repository contains the binary weights, metadata, license, and tokenizer assets for the esp32-sandbox project (projects/s3-tiny-stories). - README.md - Model Card and documentation - LICENSE - Apache 2.0 License - metadata.json - Hardware, quantization, and model architecture metadata -.bin - Compiled INT4 model weights - tokenizer.json - SentencePiece / BPE vocabulary configuration Download the binary file (model.bin) and flash it directly to your ESP32-S3: Refer to the esp32-sandbox repository for firmware building, flashing, and serial monitoring.

This model is a fine-tuned version of francesca9805/ppt-wc-loglinear-newlex-eus-before-100mb-packed-bfdisoseed3407. It has been trained using TRL. This model was trained with SFT.

This model is a fine-tuned version of francesca9805/eus-latn-100mb-ppt-mp-struct-100mbseed10. It has been trained using TRL. This model was trained with SFT.

ONNX conversions of two published line-drawing generators. They are used in the browser by the portrait pen-path tool at https://everyrobottools.com/tools/portrait with onnxruntime-web, on the WASM CPU backend. No images are uploaded; inference runs on the user's device. These files are conversions only. All credit for the models goes to the original authors listed below. Conversion scripts and the comparison against PyTorch are in the site repository under scripts/models/. All files store weights in float16. Inputs and outputs stay float32 (keepiotypes). - informative-drawings style 2 - Input input: [1,3,H,W], RGB in 0..1, no normalization. H and W are dynamic; multiples of 8 are…

Model · Text generation

evilguy-3-gguf

Alkon Alkonsafe Safe

evilguy 3 is the current version of evilguy, a fine-tune of Qwen3 14B with a specific personality: lazy, rude, moody, good at math, fluent in internet slang and memes, and unwilling to make things up. It is not meant to be a helpful assistant. It is meant to be a character. I built it by writing and generating about 7,900 conversation examples and training a LoRA on top of Qwen3 14B using Unsloth. Qwen3 runs with thinking disabled, so replies stay short instead of turning into long reasoning. The result runs locally in llama.cpp as a GGUF file. - evilguy 3 (this one): Qwen3 14B. Thinking disabled, about 7,900 examples, tool calling, slang and meme knowledge, and stronger handling of…