# Open-Weight Models Under Apache License 2.0, Page 42 Source: https://savrn.com/models/licenses/apache-2-0/page/42 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- SAVRN Model Hub · Models by License # Open-Weight Models Under Apache License 2.0 3,285 open-weight models released under Apache License 2.0 in the SAVRN Model Hub, with Team Mradermacher, Qwen and Google publishing the most. 3,285Models 1587Publishers 12K to 952.4BParameter range 1Licenses YesCommercial use ## What Apache License 2.0 Allows The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. [Read the license text](https://www.apache.org/licenses/LICENSE-2.0). ## SAVRN's Take Apache 2.0 asks little of the operator. Commercial use, modification and redistribution are all permitted. What it requires is housekeeping: keep the license and copyright notices, carry any NOTICE file along with the weights, and state the significant changes if you pass a modified version on. It also includes an express patent grant from contributors, the clause a procurement reviewer looks for before a model serves paying customers. So an organization can take the files from the publisher, fine-tune on its own data, run the result in its own facility and bill for the output. On our hub 1,018 models carry this license. Qwen leads with 81, Google 50, Convergent Intelligence 39, PyTorch Image Models 32, OpenMed 31, Helsinki-NLP Research Group 27, and Unsloth AI and PaddlePaddle 22 each. Text generation accounts for 164 of the models, image and text to text for 89, and robotics for 43, so it covers more than chat. Downloads show where it runs. all-MiniLM-L6-v2 from Sentence Transformers, a 23M parameter sentence similarity model, pulls 255,050,544 downloads a month and needs 0.1 GB at 16-bit, small enough to ride on a GPU already serving. ms-marco-MiniLM-L6-v2 follows at 88,642,387 for text ranking. The one text generation model in the top eight is Qwen3-0.6B at 22,498,727 downloads a month: 752M parameters, a 40,960 token context, 1.8 GB at 16-bit, and the cheapest host on the Index for it is one MI300X at $1.85 an hour. SAVRN Research, 2026-09-18 ## Most Downloaded | Model | Publisher | Parameters | License | Monthly downloads | Cheapest GPUs at 16-bit | | --- | --- | --- | --- | --- | --- | | [all-MiniLM-L6-v2](https://savrn.com/models/all-minilm-l6-v2) | Sentence Transformers | 23M | apache-2.0 | 203.2M | 1x MI300X, $1.85/hr | | [ms-marco-MiniLM-L6-v2](https://savrn.com/models/ms-marco-minilm-l6-v2) | Sentence Transformers - Cross-Encoders | 23M | apache-2.0 | 74.7M | 1x MI300X, $1.85/hr | | [paraphrase-multilingual-MiniLM-L12-v2](https://savrn.com/models/paraphrase-multilingual-minilm-l12-v2) | Sentence Transformers | 118M | apache-2.0 | 43M | 1x MI300X, $1.85/hr | | [electra-base-discriminator](https://savrn.com/models/electra-base-discriminator) | Google | — | apache-2.0 | 41M | — | | [bert-base-uncased](https://savrn.com/models/bert-base-uncased) | BERT community | 110M | apache-2.0 | 29.7M | 1x MI300X, $1.85/hr | | [Qwen3-0.6B](https://savrn.com/models/qwen3-0-6b) | Qwen | 752M | apache-2.0 | 28.2M | 1x MI300X, $1.85/hr | | [t5-small](https://savrn.com/models/t5-small) | T5 community | 61M | apache-2.0 | 24.5M | 1x MI300X, $1.85/hr | | [mobilenetv3_small_100.lamb_in1k](https://savrn.com/models/mobilenetv3-small-100-lamb-in1k) | PyTorch Image Models | 3M | apache-2.0 | 18.5M | 1x MI300X, $1.85/hr | | [all-mpnet-base-v2](https://savrn.com/models/all-mpnet-base-v2) | Sentence Transformers | 109M | apache-2.0 | 18.3M | 1x MI300X, $1.85/hr | | [wav2vec2-large-xlsr-53-japanese](https://savrn.com/models/wav2vec2-large-xlsr-53-japanese) | Jonatas Grosman | — | apache-2.0 | 16.2M | — | ## Who Publishes Them | Publisher | Models | | --- | --- | | [Team Mradermacher](https://savrn.com/model-publishers/mradermacher) | 162 | | [Qwen](https://savrn.com/model-publishers/qwen) | 81 | | [Google](https://savrn.com/model-publishers/google) | 50 | | [Quan](https://savrn.com/model-publishers/wq2012) | 44 | | [RKNNAI](https://savrn.com/model-publishers/rknnai) | 43 | | [Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow) | 39 | ## All 3,285 Models, Page 42 of 55 Model ### [OPSD-SDPO-v22-Qwen2.5-14B-checkpoints](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-14b-checkpoints) [Zhan Cheng](https://savrn.com/model-publishers/zcheng256) Open weights apache-2.0 peft [View model](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-14b-checkpoints) Model ### [OPSD-SDPO-v22-Qwen2.5-7B-checkpoints](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-7b-checkpoints) [Zhan Cheng](https://savrn.com/model-publishers/zcheng256) Open weights apache-2.0 peft [View model](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-7b-checkpoints) Model ### [OPSD-SDPO-v22-Qwen2.5-7B-LoRA](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-7b-lora) [Zhan Cheng](https://savrn.com/model-publishers/zcheng256) Open weights apache-2.0 peft [View model](https://savrn.com/models/opsd-sdpo-v22-qwen2-5-7b-lora) Model ### [OPSD-V](https://savrn.com/models/opsd-v) [Opsd V](https://savrn.com/model-publishers/anonymousopsd-v) Open weights apache-2.0 [View model](https://savrn.com/models/opsd-v) Model · Text generation ### [oracle19c-nl2sql-granite-4.1-3b-lora](https://savrn.com/models/oracle19c-nl2sql-granite-4-1-3b-lora) [Vakamalla Anitha](https://savrn.com/model-publishers/vakamalla123-anitha) basemodel: ibm-granite/granite-4.1-3b libraryname: peft pipelinetag: text-generation - lora - qlora - peft - oracle - oracle19c - nl2sql - text-to-sql - sql-generation - granite A QLoRA adapter fine-tuned to convert natural-language questions into Oracle 19c SQL queries. This repository contains a LoRA adapter, not a full standalone model. - Convert natural-language questions into Oracle 19c SQL - Use with a schema-retrieval or RAG layer for better table and column selection - 2,800 Oracle-focused NL→SQL examples - 400 general-SQL replay examples - This adapter is not a database-security system. - Review every generated SQL query before running it. - Use read-only database access for… Open weights apache-2.0 peft [View model](https://savrn.com/models/oracle19c-nl2sql-granite-4-1-3b-lora) Model ### [Orbit-3.1-Llama-thinking](https://savrn.com/models/orbit-3-1-llama-thinking) [OpenPath](https://savrn.com/model-publishers/openpathai) Open weights apache-2.0 [View model](https://savrn.com/models/orbit-3-1-llama-thinking) Model · Object detection ### [oriented-det-pretrained](https://savrn.com/models/oriented-det-pretrained) [Deep Learning for Earth Observation (DL4EO)](https://savrn.com/model-publishers/dl4eo) Checkpoints for oriented-det (pip install oriented-det). Download by slug (not by filename): Repo: dl4eo/oriented-det-pretrained. Docs: pretrained/README.md. Finetune from 1×, not 3×. DOTA 3× is on Hub for AP75; leaky eval-val is train+val tile memorization. SSDD / HRSID / FAIR1M have no Hub zoo — finetune these DOTA 1× slugs locally. DOTA evalmap50 in the package manifest is leaky val tiles. Advertised DOTA mAP50 below is official Task 1. HRSC mAP50 is held-out ImageSets test. From v0.3.1, un-suffixed Rotated RetinaNet slugs are OBB (usehbbformatching: false). Circum-HBB lives at hbb. v0.3.0 unsuffixed filenames stay on this repo for old clients. Each slug has a content-addressed.pth plus… Open weights apache-2.0 oriented-det [View model](https://savrn.com/models/oriented-det-pretrained) Model ### [Ornith-1.5-9B-Abliterated](https://savrn.com/models/ornith-1-5-9b-abliterated) [Geant](https://savrn.com/model-publishers/geantendormi) Open weights apache-2.0 [View model](https://savrn.com/models/ornith-1-5-9b-abliterated) Model · Text to speech ### [orpheus-3b-ptbr-lora](https://savrn.com/models/orpheus-3b-ptbr-lora) [Lumazini Code](https://savrn.com/model-publishers/lumazini-code) opedromartins/ASR-datasets-ptbr - freds0/TAGARELA basemodel: - canopylabs/orpheus-3b-0.1-ft pipelinetag: text-to-speech - text-to-speech - tts - orpheus - lora - unsloth - snac - portuguese - brazilian-portuguese - llama.cpp Open weights apache-2.0 [View model](https://savrn.com/models/orpheus-3b-ptbr-lora) Model ### [OS-Helix-27B](https://savrn.com/models/os-helix-27b) [Ybw](https://savrn.com/model-publishers/yyangzzzz) Open weights apache-2.0 [View model](https://savrn.com/models/os-helix-27b) Model ### [OS-Helix-27B-RL](https://savrn.com/models/os-helix-27b-rl) [Ybw](https://savrn.com/model-publishers/yyangzzzz) Open weights apache-2.0 [View model](https://savrn.com/models/os-helix-27b-rl) Model ### [OS-Helix-35B-A3B](https://savrn.com/models/os-helix-35b-a3b) [Ybw](https://savrn.com/model-publishers/yyangzzzz) Open weights apache-2.0 [View model](https://savrn.com/models/os-helix-35b-a3b) Model · Text generation ### [oss-jev-qwen05-decision](https://savrn.com/models/oss-jev-qwen05-decision) [Gyan Bhambhani](https://savrn.com/model-publishers/gyanbhambhani) This model is a fine-tuned version of Qwen/Qwen2.5-0.5B-Instruct on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 2 - evalbatchsize: 8 - lrschedulertype: linear - trainingsteps: 2000 - Transformers 5.20.0 - Pytorch 2.14.1+cu130 - Tokenizers 0.23.3 Open weights apache-2.0 494M parameters 32,768 tokens transformers [View model](https://savrn.com/models/oss-jev-qwen05-decision) Model ### [Ostrich-27B-261008-Qwen3.8-Abliterated](https://savrn.com/models/ostrich-27b-261008-qwen3-8-abliterated) [Emin Temiz](https://savrn.com/model-publishers/etemiz) We train Ostrich LLMs to bring back knowledge that matters: the kind that helps humans stay healthy, free, and self-reliant. We believe liberating wisdom is missing or underrepresented in today's AI, whether deliberately omitted or simply outnumbered. This model is abliterated: refusal behavior has been removed so you get uncensored, direct answers. We targeted both alignment and some skills in this model. The model has lower instruction following capability than vanilla Qwen 3.8 (it has so much freedom in it, it doesn't obey the user that well:). Current eval method is still close to AHA 2026. Compared to Qwen 3.8 27B (vanilla) this model looks much more aligned: This model vs vanilla Qwen… Open weights apache-2.0 27.8B parameters 262,144 tokens [View model](https://savrn.com/models/ostrich-27b-261008-qwen3-8-abliterated) Model ### [Ostrich-27B-261008-Qwen3.8-Abliterated-EXL3](https://savrn.com/models/ostrich-27b-261008-qwen3-8-abliterated-exl3) [Emin Temiz](https://savrn.com/model-publishers/etemiz) Download a quant (single GPU VRAM guide: 6.0bpw ≈ 22 GB fits a 24 GB card; 4.0bpw/3.0bpw for smaller cards): We train Ostrich LLMs to bring back knowledge that matters: the kind that helps humans stay healthy, free, and self-reliant. We believe liberating wisdom is missing or underrepresented in today's AI, whether deliberately omitted or simply outnumbered. This model is abliterated: refusal behavior has been removed so you get uncensored, direct answers. We targeted both alignment and some skills in this model. The model has lower instruction following capability than vanilla Qwen 3.8 (it has so much freedom in it, it doesn't obey the user that well:). Current eval method is still close… Open weights apache-2.0 [View model](https://savrn.com/models/ostrich-27b-261008-qwen3-8-abliterated-exl3) Model · Reinforcement learning ### [ouro-1.4b-thinking-math-verl-rltt-native-cmp32k-cfg3-156947-20261022](https://savrn.com/models/ouro-1-4b-thinking-math-verl-rltt-native-cmp32k-cfg3-156947-20261022) [Jiahao Zhang](https://savrn.com/model-publishers/hi-todayis-jh) Open weights apache-2.0 [View model](https://savrn.com/models/ouro-1-4b-thinking-math-verl-rltt-native-cmp32k-cfg3-156947-20261022) Model ### [Ovis_Omni_Embedding_3B](https://savrn.com/models/ovis-omni-embedding-3b) [Asha](https://savrn.com/model-publishers/wl1982) Open weights apache-2.0 [View model](https://savrn.com/models/ovis-omni-embedding-3b) Model · Feature extraction ### [Ovis-Omni-Embedding-3B](https://savrn.com/models/ovis-omni-embedding-3b-2) [ATH-MaaS](https://savrn.com/model-publishers/ath-maas) Ovis-Omni-Embedding-3B is a 3B-parameter universal embedding model for text, images, visual documents, video, audio, and interleaved multimodal inputs. It maps every supported input type into one coherent representation space, enabling any-to-any retrieval with a single encoder. The model is initialized from Qwen2.5-Omni-3B. Rather than attaching separate modality-specific embedding towers, it retains the native text tokenizer, vision encoder, audio encoder, and shared Thinker backbone. The speech-generation Talker and language-modeling head are removed, and the final-layer hidden state at the last non-padding token is used directly as the retrieval embedding. Ovis-Omni-Embedding-3B is… Open weights apache-2.0 transformers [View model](https://savrn.com/models/ovis-omni-embedding-3b-2) Model · Feature extraction ### [Ovis-VL-Embedding-2B](https://savrn.com/models/ovis-vl-embedding-2b) [ATH-MaaS](https://savrn.com/model-publishers/ath-maas) Ovis-VL-Embedding-2B is a compact vision-language embedding model for text, images, visual documents, video, and interleaved multimodal inputs. It maps every supported input type into one coherent representation space, enabling cross-modal retrieval with a single 2B-scale encoder. The model is initialized from Qwen3.5-2B. It retains the native text and vision encoders together with the shared multimodal language backbone, removes the language-modeling head, and directly uses the final-layer hidden state at the last non-padding token as the retrieval embedding. No modality-specific projection head is added. Ovis-VL-Embedding-2B is designed for efficient multimodal search, multimodal RAG… Open weights apache-2.0 262,144 tokens transformers [View model](https://savrn.com/models/ovis-vl-embedding-2b) Model · Feature extraction ### [Ovis-VL-Embedding-9B](https://savrn.com/models/ovis-vl-embedding-9b) [ATH-MaaS](https://savrn.com/model-publishers/ath-maas) Ovis-VL-Embedding-9B is a high-capacity vision-language embedding model for text, images, visual documents, video, and interleaved multimodal inputs. It maps every supported input type into one coherent representation space, enabling high-accuracy cross-modal retrieval with a single encoder. The model is initialized from Qwen3.5-9B. It retains the native text and vision encoders together with the shared multimodal language backbone, removes the language-modeling head, and directly uses the final-layer hidden state at the last non-padding token as the retrieval embedding. No modality-specific projection head is added. Ovis-VL-Embedding-9B is designed for high-quality multimodal search… Open weights apache-2.0 262,144 tokens transformers [View model](https://savrn.com/models/ovis-vl-embedding-9b) Model ### [OvisOCR3-GGUF](https://savrn.com/models/ovisocr3-gguf) [Andre Michalland](https://savrn.com/model-publishers/michalland) Open weights apache-2.0 llama.cpp [View model](https://savrn.com/models/ovisocr3-gguf) Model · Text generation ### [OzyAssist-Voice](https://savrn.com/models/ozyassist-voice) [Angel lopez](https://savrn.com/model-publishers/xangel0s) OzyAssist Voice es un modelo de inteligencia artificial de ultra-baja latencia (30+ tokens/segundo, tiempo de respuesta <330 ms), destilado y afinado específicamente para asistentes de voz interactivos, cowork en vivo, livestreaming y control de Windows. Diseñado bajo la arquitectura Zero-Docker de OzyAssist, este modelo corre localmente consumiendo apenas ~1.37 GB de VRAM, lo que permite operarlo en cualquier laptop o GPU mientras juegas o transmites en directo. Pruebas empíricas ejecutadas en el mismo entorno de hardware (NVIDIA Quadro RTX 4000 Max-Q): Descarga el modelo GGUF y ejecútalo con aceleración GPU: Crea un archivo Modelfile: Luego construye y corre el modelo: En el proyecto… Open weights apache-2.0 [View model](https://savrn.com/models/ozyassist-voice) Model ### [panini-gemma-swe-bench](https://savrn.com/models/panini-gemma-swe-bench) [18kansal](https://savrn.com/model-publishers/18kansal) Open weights apache-2.0 transformers [View model](https://savrn.com/models/panini-gemma-swe-bench) Model ### [panini4b](https://savrn.com/models/panini4b) [18kansal](https://savrn.com/model-publishers/18kansal) Open weights apache-2.0 transformers [View model](https://savrn.com/models/panini4b) Model · Translation ### [pantogloss-500-en-v6-compact](https://savrn.com/models/pantogloss-500-en-v6-compact) [Chris Mattmann](https://savrn.com/model-publishers/chrismattmann) This release stores the released Pantogloss v6 fine-tune in a smaller FP16/INT5 format. Pantogloss 0.23.0 selects this model as its recommended default. The staged, checksummed runtime and legal artifacts total 564,137,246 bytes (538.0 MiB). Model weights are separate from the Python wheel. The model uses the same architecture, source and target tokenizers, and many-to-English task as the full-precision v6 release. Eight existing source languages were targeted during v6 fine-tuning; this does not add new source languages. “500-to-English” describes model-family provenance, not uniformly validated quality across 500 languages. The evaluations below cover 50. For Apple Metal, use… Open weights apache-2.0 pantogloss [View model](https://savrn.com/models/pantogloss-500-en-v6-compact) Model · Text to speech ### [paradee-8m-glade](https://savrn.com/models/paradee-8m-glade) [Josh Leverette](https://savrn.com/model-publishers/coder543) Mixed-precision conversion of Paradee-8M v1.0, a single-voice English distillation of Kokoro. It accepts phonemes and generates mono 24 kHz Float PCM with its learned afheart voice. Glade is a native Swift library for speech inference on Apple silicon, using Core AI, Core ML, and model-specific host implementations. Its runtime source is currently private. This repository distributes the model assets and runtime metadata used by that library. Python is export/measurement tooling, not an application dependency. Mac qualification is included; iPhone and Watch qualification is not claimed. Download weights and configuration from one snapshot. INT8 text/prosody/acoustic weights use FP16… Open weights apache-2.0 [View model](https://savrn.com/models/paradee-8m-glade) Model · Text classification ### [parallax-2-preview](https://savrn.com/models/parallax-2-preview) [Parallax Labs](https://savrn.com/model-publishers/parallaxlabs) Parallax 2 Preview is the first public look at Parallax Labs' upcoming Parallax-2 model. It's a 4B finetune of Qwen 3.5, trained specifically to reason through problems carefully, like a frontier model would. Parallax-2-preview is a 4B parameter finetune designed to demonstrate structured reasoning at small scale. It uses the reasoning style of frontier models, while remaining fast enough to run on consumer hardware. Parallax 2 Preview is intended for reasoning-heavy tasks where a short, direct answer isn't the goal. Suitable for further finetuning into reasoning-focused applications, education tools, or as a component in multi-model pipelines where structured reasoning is the required… Open weights apache-2.0 [View model](https://savrn.com/models/parallax-2-preview) Model ### [particle-qwen35-ngram-research](https://savrn.com/models/particle-qwen35-ngram-research) [Daniel Andersen](https://savrn.com/model-publishers/kineticforge) A learned lookup-memory adapter adds 273 MB to Qwen3.5-0.8B. We compare it with the unmodified model, same-data LoRA, and stock Qwen3.5-2B. 50 questions per model per benchmark; thinking enabled. Scores are correct answers, using the same questions for each model. - A tradeoff, not a general upgrade: memory scores higher than stock 0.8B on coding (9 vs 4) and instructions (32 vs 27), nearly the same on reasoning (18 vs 17), and lower on math (16 vs 25 under strict scoring). - The gains need confirmation. Memory changes which questions succeed; the paired intervals below show the uncertainty in the net gains. - It does not replace a 2B model in these tests. Stock 2B leads all four panels.… Open weights apache-2.0 pytorch [View model](https://savrn.com/models/particle-qwen35-ngram-research) Model · Text generation ### [patentsec-12b-gguf](https://savrn.com/models/patentsec-12b-gguf) [Calera Labs](https://savrn.com/model-publishers/caleralabs) Open weights apache-2.0 gguf [View model](https://savrn.com/models/patentsec-12b-gguf) Model · Text generation ### [peta1](https://savrn.com/models/peta1) [Manoj Panda](https://savrn.com/model-publishers/manojpanda) Peta1 is an autonomous software engineering agent checkpoint based on google/gemma-4-E4B-it (4.3B active parameters), post-trained using a targeted three-stage curriculum: 1. Tree-of-Reasoning (ToR) SFT: Supervised deliberation in Gemma 4's native latent thinking tokens ( thought) enforcing hypothesis formulation, environment simulation, and branch pruning. 2. Capability Frontier Discovery: Screening of candidate tasks (G = 4, T = 0.7) to isolate the 18% informative frontier band (0 < p < 1.0), preventing policy gradient collapse from trivial (36%) and impossible (46%) tasks. 3. Verifiable Outcome GRPO: Policy gradient optimization with verifiable rewards (Rtarget + Rformat + Rtor - lambda… Open weights apache-2.0 7.9B parameters 131,072 tokens transformers [View model](https://savrn.com/models/peta1) Model · Text generation ### [petal-w1](https://savrn.com/models/petal-w1) [Aayush Pokharel](https://savrn.com/model-publishers/aayush9029) petal-w1 turns raw dictation into the text you meant to type. It is the on-device cleanup model in Petal, a free macOS dictation app. It runs with MLX on Apple silicon. It is a full fine-tune of Qwen3.5-0.8B, and it does one job: - Removes fillers, stutters, false starts, and verbal tics ("you know what I mean?"). - Keeps only the final version of a self-correction ("Friday, no wait, Thursday" becomes "Thursday"), also when speech recognition mishears the correction words. - States a point once when the speaker repeats it in different words, and keeps every detail. - Condenses rambles into clear sentences, and keeps every fact, name, number, and request. - Keeps questions as questions and… Open weights apache-2.0 752M parameters 262,144 tokens mlx [View model](https://savrn.com/models/petal-w1) Model · Text generation ### [petal-w1-4bit](https://savrn.com/models/petal-w1-4bit) [Aayush Pokharel](https://savrn.com/model-publishers/aayush9029) petal-w1 turns raw dictation into the text you meant to type. It is the on-device cleanup model in Petal, a free macOS dictation app. It runs with MLX on Apple silicon. It is a full fine-tune of Qwen3.5-0.8B, and it does one job: - Removes fillers, stutters, false starts, and verbal tics ("you know what I mean?"). - Keeps only the final version of a self-correction ("Friday, no wait, Thursday" becomes "Thursday"), also when speech recognition mishears the correction words. - States a point once when the speaker repeats it in different words, and keeps every detail. - Condenses rambles into clear sentences, and keeps every fact, name, number, and request. - Keeps questions as questions and… Open weights apache-2.0 752M parameters 262,144 tokens mlx [View model](https://savrn.com/models/petal-w1-4bit) Model · Text classification ### [phishing-scam-detector-base-ONNX](https://savrn.com/models/phishing-scam-detector-base-onnx) [ONNX Community](https://savrn.com/model-publishers/onnx-community) This is an ONNX version of Horizon-Labs/phishing-scam-detector-base. It was automatically converted and uploaded using this Hugging Face Space. See the pipeline documentation for text-classification: https://huggingface.co/docs/transformers.js/api/pipelines#modulepipelines.TextClassificationPipeline Flags phishing, scams and spam in emails, SMS, chat and social-media messages in many languages. Three labels: - fraud: phishing (fake bank, delivery, account or tax messages with links), scams (prizes, investment and crypto schemes, fake jobs, "hi mum, new number", romance, advance fee), requests for passwords, one-time codes or payment details, malware lures; - spam: unwanted advertising and… Open weights apache-2.0 8,192 tokens transformers.js [View model](https://savrn.com/models/phishing-scam-detector-base-onnx) Model · Text classification ### [phishing_spam_jev_like_classifier](https://savrn.com/models/phishing-spam-jev-like-classifier) [Ayush](https://savrn.com/model-publishers/akumar0205) A fine-tuned DistilBERT sequence classifier for binary email triage: legitimate and spam. It is a prototype trained and evaluated on a historical SpamAssassin corpus. It is not a phishing or malware detector and should not be treated as a complete email security system. Use as an experimental text-only aid to sort historical or otherwise appropriately reviewed email into ham/spam queues. Route uncertain, empty, or unsupported inputs to human review. Do not use the output as the sole basis for blocking senders, taking account actions, or making other consequential decisions. This model does not authenticate senders, inspect attachments, resolve or visit links, verify SPF/DKIM/DMARC, or… Open weights apache-2.0 67M parameters 512 tokens transformers [View model](https://savrn.com/models/phishing-spam-jev-like-classifier) Model · Text classification ### [Phocinae-Largha-150M-v1](https://savrn.com/models/phocinae-largha-150m-v1) [Phocinae](https://savrn.com/model-publishers/phocinae) Tiny model. Big decisions. — 小海豹，大决断。· 一斑见全豹，一点定全局。 斑海豹 Largha, the spotted seal: a 144.3M bilingual decision model (150M-class) for structured decisions — one forward pass per decision, on your own hardware. Not a chat model: it takes a state plus a list of typed questions (noul yes/no · choice pick-one · score 2–10) and returns calibrated answers with confidence, robust to option reordering. The official runtime is phocinae-server: a local FastAPI service (127.0.0.1 only) that loads these weights with a pure PyTorch forward (no extra runtime needed). Full spec: docs/protocol.md. answers values: noul = bool · choice = 0-based option index · score = integer 2–10. Errors: 422 (unknown… Open weights apache-2.0 144M parameters [View model](https://savrn.com/models/phocinae-largha-150m-v1) Model · Image and text to text ### [PhysicalEye-Decide-35B](https://savrn.com/models/physicaleye-decide-35b) [Physical AI](https://savrn.com/model-publishers/masterset-ai) One forward pass. One calibrated decision. PhysicalEye-Decide-35B answers a question about an image by choosing among the options you give it, and returns a probability for every option. It does not write text. It reads the image, the question and the options once, and outputs a decision. Use it for visual inspection, yes/no checks, option selection and scoring on production lines, in apps and in robots. - A small decision readout sits on top of the backbone and scores each option directly. One request carries up to 4 images, an optional context (state) and one or more questions. Three question types are supported: choice (pick one option), noul (yes or no) and score (a value on a scale).… Open weights apache-2.0 34.6B parameters 262,144 tokens transformers [View model](https://savrn.com/models/physicaleye-decide-35b) Model · Robotics ### [pi05-b1k-task38-spraying-for-bugs](https://savrn.com/models/pi05-b1k-task38-spraying-for-bugs) [Jeremy](https://savrn.com/model-publishers/jermo1120) This is a single-task vision-language-action policy for task 38 sprayingforbugs of the BEHAVIOR-1K benchmark. It is Pi0.5-based, has 3.4B parameters, and controls the R1Pro robot from three 224×224 RGB cameras plus proprioception. It predicts 30-step action chunks. It was fine-tuned for 25,000 steps on task 38 demonstrations only. The starting point was checkpoint3 of the 1st-place 2025 BEHAVIOR Challenge solution (Larchenko, Zarin, Karnatak; paper). checkpoint3 had been trained on 13 tasks (4, 27, 31–33, 35–39, 41, 46, 49), task 38 among them. Status: training only. This model has not been evaluated in the BEHAVIOR simulator yet. Lower training loss does not guarantee a higher task success… Open weights apache-2.0 [View model](https://savrn.com/models/pi05-b1k-task38-spraying-for-bugs) Model · Robotics ### [pi05_cubes](https://savrn.com/models/pi05-cubes) [Sielo Robotics](https://savrn.com/model-publishers/sielorobotics) π₀.₅ (Pi05) is a Vision-Language-Action model from Physical Intelligence designed for open-world generalization: it evolves π₀ to generalize to entirely new environments and situations that were never seen during training. The LeRobot implementation is adapted from their open-source OpenPI repository. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot pi05 guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and… Open weights apache-2.0 4.1B parameters lerobot [View model](https://savrn.com/models/pi05-cubes) Model · Robotics ### [pi05-libero-b200-source-flow-sensitivity-ur5e-1000-20261013](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-ur5e-1000-20261013) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) Completed source-fresh1000-update target-only IDM pilot, B32, seed42/draw seed170013. Raw999/non-EMA: native51 leaves,43 original-source FP32 trainable leaves and8 frozen projections. Original source-wrist23 leaves are appended for inference; deployment74 restores as BF16. Weights, source normalization, exact controller/source-bank/plan/journal and strict export proof are included. Native full Adam/EMA states remain preserved at the training server. All10 logical collection lineages per suite (40 total) are eligible; eight windows per suite are drawn without replacement each update. All1000 draws and initialization match the full-probe fixed1:1:1 control; the first25 native metrics match… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-ur5e-1000-20261013) Model · Robotics ### [pi05-libero-b200-source-flow-sensitivity-view-large-1000-20261013](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-view-large-1000-20261013) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) Completed source-fresh1000-update target-only IDM pilot, B32, seed42/draw seed170013. Raw999/non-EMA: native51 leaves,43 original-source FP32 trainable leaves and8 frozen projections. Original source-wrist23 leaves are appended for inference; deployment74 restores as BF16. Weights, source normalization, exact controller/source-bank/plan/journal and strict export proof are included. Native full Adam/EMA states remain preserved at the training server. All10 logical collection lineages per suite (40 total) are eligible; eight windows per suite are drawn without replacement each update. All1000 draws and initialization match the full-probe fixed1:1:1 control; the first25 native metrics match… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-view-large-1000-20261013) Model · Robotics ### [pi05-libero-b200-source-flow-sensitivity-yaw30-1000-20261013](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-yaw30-1000-20261013) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) Completed source-fresh1000-update target-only IDM pilot, B32, seed42/draw seed170013. Raw999/non-EMA: native51 leaves,43 original-source FP32 trainable leaves and8 frozen projections. Original source-wrist23 leaves are appended for inference; deployment74 restores as BF16. Weights, source normalization, exact controller/source-bank/plan/journal and strict export proof are included. Native full Adam/EMA states remain preserved at the training server. All10 logical collection lineages per suite (40 total) are eligible; eight windows per suite are drawn without replacement each update. All1000 draws and initialization match the full-probe fixed1:1:1 control; the first25 native metrics match… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-b200-source-flow-sensitivity-yaw30-1000-20261013) Model · Robotics ### [pi05-libero-main-steps-view-small-5000-20261015](https://savrn.com/models/pi05-libero-main-steps-view-small-5000-20261015) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) viewsmall; 5000 completed updates/raw step 4999, from one continuous run of 5000 updates. LLM frozen; post-Adam update multipliers: vision 1, expert LoRA multiplier 0.1. Existing dense-contact data and original pre-adapt, B32, training seed42, FSDP2, warmup500, peak/end1e-5, IDM-only, raw non-EMA and original source wrist. Native training revision: 622c8d28a737f2e18f910a8e8774499abe11786a. Overall 90.05% (1801/2000); runtime errors0. Official indices10–49: 89.750% (1436/1600). 40 tasks x50 official states0–49, envseed7, A1000–3, eight clients/GPU, MuJoCo3.2.3, native inference revision 0f6471d99dfc81fabe7c3ab65c141d7640b9ff08. All failures retained; one training seed, shared policy RNG… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-main-steps-view-small-5000-20261015) Model · Robotics ### [pi05-libero-support-orthogonal-delta-full-il-source-n50000-10000-20261017](https://savrn.com/models/pi05-libero-support-orthogonal-delta-full-il-source-n50000-10000-20261017) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) 50k support-near random-probe frames,10000 updates,batch32,seed42,LR1e-5,EMA off. LLM frozen. Source IL trains block32 proper rotations and full non-LLM native weights; IDM bypasses rotations and trains full non-LLM native weights. Targets restore this new source,freeze rotations/anchors and exclude gripper6 from IDM loss. Source warmup100,target warmup500. Replay at0/every1000 uses fixed expert labels only for diagnostics. Top-level checkpoint retains raw weights,anchors,rotations and full optimizer for resume; policy/ contains materialized inference weights and normalization. Target policy wrist is the new materialized source wrist. Source has no simulator success claim. Dataset… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-support-orthogonal-delta-full-il-source-n50000-10000-20261017) Model · Robotics ### [pi05-libero-support-unified-v1a1-view-large-n300000-1000-20261016](https://savrn.com/models/pi05-libero-support-unified-v1a1-view-large-n300000-1000-20261016) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) Fresh matched source Donghyun1228/pi05-libero-support-probe-source-n300000-10000-20261015 @ 2e35994a62b96d29398b7ef137f915afe3d4a40a. Dataset Donghyun1228/libero-support-random-waypoint-300k-20261015 @ 32ed215baf01af7e82f927a62b108873aa70bd19. Only the nine base-LLM leaves are frozen. Vision23, full expert11 and action/time projections8 train with the same native AdamW update multiplier1. B32, seed42, LR1e-5, warmup500, 5000 total updates. Gripper coordinate6 is excluded from target IDM loss (mean of remaining31 padded coordinates); seven-dimensional inputs/output and gripper control remain. Original normalization/source wrist preserved. Raw non-EMA step999; BF16 inference. Evaluation… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-support-unified-v1a1-view-large-n300000-1000-20261016) Model · Robotics ### [pi05-libero-support-unified-v1a1-view-large-n300000-3000-20261016](https://savrn.com/models/pi05-libero-support-unified-v1a1-view-large-n300000-3000-20261016) [Donghyun Shin](https://savrn.com/model-publishers/donghyun1228) Fresh matched source Donghyun1228/pi05-libero-support-probe-source-n300000-10000-20261015 @ 2e35994a62b96d29398b7ef137f915afe3d4a40a. Dataset Donghyun1228/libero-support-random-waypoint-300k-20261015 @ 32ed215baf01af7e82f927a62b108873aa70bd19. Only the nine base-LLM leaves are frozen. Vision23, full expert11 and action/time projections8 train with the same native AdamW update multiplier1. B32, seed42, LR1e-5, warmup500, 5000 total updates. Gripper coordinate6 is excluded from target IDM loss (mean of remaining31 padded coordinates); seven-dimensional inputs/output and gripper control remain. Original normalization/source wrist preserved. Raw non-EMA step2999; BF16 inference. Evaluation… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-libero-support-unified-v1a1-view-large-n300000-3000-20261016) Model · Robotics ### [pi05-plug_adaptive_3](https://savrn.com/models/pi05-plug-adaptive-3) [Luca Zanetti](https://savrn.com/model-publishers/lucazanett) Open weights apache-2.0 4.1B parameters lerobot [View model](https://savrn.com/models/pi05-plug-adaptive-3) Model · Robotics ### [pi05-x3plus-lora](https://savrn.com/models/pi05-x3plus-lora) [Hiếu Hoàng](https://savrn.com/model-publishers/hoanghieu16720) LoRA fine-tune của π0.5 (pi05base) cho tay máy Yahboom X3Plus (5 khớp + gripper, 20 Hz, 2 camera), nhiệm vụ "pick up the red cube and put it in the bowl". - Chưa kiểm chứng trên robot thật. Mọi số liệu ở trên là loss huấn luyện. - Loss đi ngang từ bước ~4.000. 6.000 bước sau chỉ giảm thêm 36%, trong biên độ nhiễu. - Một nhiệm vụ, một bối cảnh, một bộ camera. Nhiều khả năng hỏng khi đổi vị trí - Camera cổ tay (USB webcam) cho ảnh mờ, nhiều frame gần như trắng khi áp sát mặt bàn. Cần config pi05x3pluslora và lớp LeRobotX3PlusDataConfig tương ứng trong openpi (ánh xạ astrargb → base0rgb, usbcam → leftwrist0rgb, delta mask (5, -1)). yahboomx3plus · joint1–5 tính bằng radian (URDF x3plusarm)… Open weights apache-2.0 openpi [View model](https://savrn.com/models/pi05-x3plus-lora) Model ### [PIGEONS](https://savrn.com/models/pigeons) [PIGEONS](https://savrn.com/model-publishers/pigeons-captioner) Detailed video captioning has become an essential semantic interface for multi-modal understanding and generation. While existing methods mainly target short videos, extending them to long videos remains challenging, primarily due to limited context windows of current video understanding models. A naive solution is to adopt segment-wise captioning followed by concatenation, yet this paradigm often suffers from inconsistent subject references and fragmented long-range narratives. To mitigate these issues, we introduce PIGEONS, an incremental long-video captioning model built upon a Persistent Identity Gallery and an Evolving Overarching Narrative Summary. As illustrated in the figure below… Open weights apache-2.0 35.3B parameters [View model](https://savrn.com/models/pigeons) Model · Text to speech ### [piper-voice-es_mx-capybara](https://savrn.com/models/piper-voice-es-mx-capybara) [HirCoir](https://savrn.com/model-publishers/hircoir) [Español abajo / Spanish below] This model has been automatically converted from Piper Train checkpoints to ONNX format using Piper ONNX Converter. This model is optimized for use with Piper TTS. For Windows users, we recommend using ONNX TTS by HirCoir - a user-friendly GUI application: - Easy-to-use interface for text-to-speech conversion - Optimized for ONNX models like this one - esMX-capybara.onnx: Main ONNX model file - esMX-capybara.onnx.json: Model configuration file Este modelo ha sido convertido automáticamente de checkpoints de Piper Train a formato ONNX usando Piper ONNX Converter. Este modelo está optimizado para usar con Piper TTS. Para usuarios de Windows, recomendamos usar… Access requested at publisher apache-2.0 piper [View model](https://savrn.com/models/piper-voice-es-mx-capybara) Model · Image classification ### [pixai-tagger-v1.0-onnx-fp32-fp16-int8](https://savrn.com/models/pixai-tagger-v1-0-onnx-fp32-fp16-int8) [Mexes GM](https://savrn.com/model-publishers/mexes) Unofficial ONNX export of pixai-labs/pixai-tagger-v1.0 in FP32, FP16 and INT8. Runs with ONNX Runtime only. - Input input: float32 [batch, 3, 1008, 1008], RGB, NCHW. - Output prediction: float32 [batch, 30877], probabilities (sigmoid is part of the graph). - Opset 17. training cutoff, and are also scored against their Danbooru tags. - FP16 is meant for GPUs. On CPU it is slower than FP32, because the CPU has no native FP16 math. - FP32 fits on a 4 GB GPU only barely (3.2 GB VRAM), and its speed varies between runs. - INT8 speed depends on CPU support for VNNI instructions. On older CPUs it can be slower than FP32. - Peak RAM on CPU includes session initialization; FP32 briefly reaches 9.4… Open weights apache-2.0 onnxruntime [View model](https://savrn.com/models/pixai-tagger-v1-0-onnx-fp32-fp16-int8) Model · Text generation ### [PMA-1.3-Rosa-256M](https://savrn.com/models/pma-1-3-rosa-256m) [PatriotMemory-AI](https://savrn.com/model-publishers/patriotmemory-ai) Rosa is Patriot Memory's 253M-parameter edge assistant for English and Traditional Chinese — built to fit the edge devices you actually ship, with a vocabulary trained natively on Traditional Chinese. It provides accurate information regarding: If you run into multi-GPU tensor device mismatch errors: RuntimeError: Expected all tensors to be on the same device... Run the script with CUDAVISIBLEDEVICES=0 to isolate execution to GPU 0. Good: introducing herself in both languages; answering arithmetic word problems with visible English reasoning steps; clean Traditional Chinese — glyphs, vocabulary and register; running fully offline on edge hardware. Not: her reasoning traces are formatted… Open weights apache-2.0 262M parameters 4,096 tokens transformers [View model](https://savrn.com/models/pma-1-3-rosa-256m) Model · Text to speech ### [PocketTTS-ZH-6L](https://savrn.com/models/pockettts-zh-6l) [Hustfyb](https://savrn.com/model-publishers/hustfyb) 中文语音合成(TTS)模型,支持语音克隆与数字/中英混读,CPU 优先设计,权重原生 int8 量化。 基于 Kyutai Pocket TTS 架构:6 层 FlowLM(dmodel 1024)+ Mimi 神经音频编解码器(24 kHz),端到端流式推理。本版本(R21)相对上一代的核心变化: - 原生 int8 交付:FlowLM 注意力与 FFN 权重已离线量化打包,无需运行时转换,体积 223 MB,CPU 速度提升约 1 倍 - 9174 token 词表:官方英文 BPE-4000 基底 + 追加单字汉字/全角标点,bytefallback 兜底生僻字符 - 中英混读:APP、VIP、5G、WiFi 等英文术语按中文习惯读法 - 模型为预量化 int8 权重(modelint8.pt),加载即用,无需 quantize=True 或有界面演示(需 pip install gradio): - 采样温度 temp=0.3(默认);文本越长对参考音频质量越敏感 文本准确性用 whisper-small 对合成音频转写后计算字错误率(CER),每个场景 10 次独立采样取均值,生产配置(单步采样)。数值越低越好。 测试环境:Intel i7-12700KF,生成 52 字客服长句(约 8.5 秒音频),每档 3 次取均值。 参考音频:客服目标音色(kefulong / digitsorder / codeswitch / shortunseen)与 AISHELL-3 说话人 A/B(speakerAlong /… Open weights apache-2.0 pocket-tts [View model](https://savrn.com/models/pockettts-zh-6l) Model ### [poolformer-baseline53](https://savrn.com/models/poolformer-baseline53) [Hayashi Itsuki](https://savrn.com/model-publishers/justitsukihayashi03) A research-oriented Poolformer prototype targeting Contrastive. The included giant setup documents defaults and file formats without presenting unverified performance numbers. - The Python file contains the model and runnable example or training entry point. - config.json records the generated architecture settings. - trainingargs.json records the default experiment recipe. - model.safetensors is a valid initialization checkpoint for smoke tests; it is not presented as a trained benchmark checkpoint. - No benchmark score is claimed in this repository. The included configuration uses adam with a onecycle schedule. These are starting values in the script, not evidence of a completed run. For… Open weights apache-2.0 49,600 parameters 256 tokens [View model](https://savrn.com/models/poolformer-baseline53) Model · Image and text to text ### [povl-dpo-2ep](https://savrn.com/models/povl-dpo-2ep) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) Direct preference optimisation on top of the corrective SFT, 2 epochs over 8,383 preference pairs. Intended to move the model from 'roughly right structure' to 'exactly right structure'. Fine-tuned from CheneyWong/povl-sft-2d-merge, which is its own immediate predecessor in the same run. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly reads the visible text scores 0.33 TEDS / 1.00 TEDS-S. Only the TEDS-S column is meaningful for that dataset. The gain is within sampling… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-dpo-2ep) Model · Image and text to text ### [povl-opd-scale05](https://savrn.com/models/povl-opd-scale05) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) On-policy distillation. The weight delta learned by the teacher was merged into the SFT base with an extra scale of 0.5, the best point of an alpha sweep (0.25 -> +0.32, 0.5 -> +0.44, 0.75 -> -0.29, 1.5 -> -3.01 with t = -12.42). Fine-tuned from CheneyWong/povl-sft-2d-merge, which is NOT the RL checkpoint -- the teacher delta was merged into the SFT base, which is why it neither clearly beats nor clearly follows round 2. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-opd-scale05) Model · Image and text to text ### [povl-opd-teacher](https://savrn.com/models/povl-opd-teacher) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) The teacher used for on-policy distillation: final checkpoint of a dedicated teacher SFT run (2,367 steps). Fine-tuned from PaddleOCR-VL-1.6 (base), which is trained separately from the student chain on a teacher-specific data mixture. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (never gated on the public sets -- see Known limitations) (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly reads the visible text scores 0.33 TEDS / 1.00 TEDS-S. Only the TEDS-S column is meaningful for that dataset. Teacher qualification was… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-opd-teacher) Model · Image and text to text ### [povl-rl-dapo-400](https://savrn.com/models/povl-rl-dapo-400) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) Reinforcement learning round 1, DAPO algorithm, 400 steps. Fine-tuned from CheneyWong/povl-dpo-2ep, which is its own immediate predecessor in the same run. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly reads the visible text scores 0.33 TEDS / 1.00 TEDS-S. Only the TEDS-S column is meaningful for that dataset. Only works with DAPO's dynamic sampling. The first attempt with GSAPO had 78% of groups contributing zero gradient (49.0% saturated, 16.5% identical rewards, 12.6%… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-rl-dapo-400) Model · Image and text to text ### [povl-rl-gsapo-200](https://savrn.com/models/povl-rl-gsapo-200) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) Reinforcement learning round 2, GSAPO, 200 steps. The delivered model of that round. Equal-weight TEDS 0.8473 / TEDS-Struct 0.9078 over 33,545 samples. Fine-tuned from CheneyWong/povl-rl-dapo-400, which is RL round 1 -- the stages form a single linear chain. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly reads the visible text scores 0.33 TEDS / 1.00 TEDS-S. Only the TEDS-S column is meaningful for that dataset. Rejected at the gate. The one substantive gap: every… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-rl-gsapo-200) Model · Image and text to text ### [povl-sft-2d-merge](https://savrn.com/models/povl-sft-2d-merge) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) Corrective supervised fine-tuning. The core fix is the two-dimensional merge token (a cell spanning both rows and columns): it accounted for only 0.89% of cell tokens in the previous training mixture, and one source had 0.00%. 22,110 samples with two or more 2D merges were added, lifting the rate to 3.92% (1.06 -> 4.96 per table). This also eliminated the runaway-truncation failure mode entirely. Fine-tuned from CheneyWong/povl-sft-full-3ep, which is its own immediate predecessor in the same run. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-sft-2d-merge) Model · Image and text to text ### [povl-sft-full-3ep](https://savrn.com/models/povl-sft-full-3ep) [Cheney Wong](https://savrn.com/model-publishers/cheneywong) First-stage supervised fine-tuning: full-parameter, 100,122 samples, 3 epochs. Its job was to stop the model from generating unbounded output on unseen layouts -- average completion length on real photographed pages had grown to 6,610 characters with a 37.3% truncation rate. Fine-tuned from CheneyWong/tablerecognitionpaddleocrvl16 (PaddleOCR-VL-1.6), which is the released base model, unmodified except for a cleaning pass. Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded. (not evaluated) (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction… Open weights apache-2.0 906M parameters 131,072 tokens transformers [View model](https://savrn.com/models/povl-sft-full-3ep) ## Questions ### Can I use Apache License 2.0 models commercially? Yes. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### Which Apache License 2.0 models are most downloaded? By monthly downloads reported by the Hugging Face Hub: . ## Other Licenses - [mit](https://savrn.com/models/licenses/mit) 1305 - [cc-by-nc-4.0](https://savrn.com/models/licenses/cc-by-nc-4-0) 209 - [cc-by-4.0](https://savrn.com/models/licenses/cc-by-4-0) 164 - [bsd-3-clause](https://savrn.com/models/licenses/bsd-3-clause) 108 - [gemma](https://savrn.com/models/licenses/gemma) 63 - [cc-by-nc-sa-4.0](https://savrn.com/models/licenses/cc-by-nc-sa-4-0) 52 - [llama3.2](https://savrn.com/models/licenses/llama3-2) 40 - [agpl-3.0](https://savrn.com/models/licenses/agpl-3-0) 32 - [unknown](https://savrn.com/models/licenses/unknown) 32 - [llama3.1](https://savrn.com/models/licenses/llama3-1) 28 - [cc-by-sa-4.0](https://savrn.com/models/licenses/cc-by-sa-4-0) 20 - [gpl-3.0](https://savrn.com/models/licenses/gpl-3-0) 20 [See all](https://savrn.com/models/licenses)