Filter by type:
Filter by tags:
streaming-zipformer-en-sherpa
Streaming English ASR: sherpa-onnx zipformer transducer (int8, chunk-16 left-128). Low-latency real-time transcription with endpoint detection via sherpa-onnx's online recognizer. English-only; for multilingual offline ASR see omnilingual-0.3b-ctc-q8-sherpa.
Repository: localaiLicense: apache-2.0
vits-ljs-sherpa
VITS-LJS English single-speaker TTS served through the sherpa-onnx backend. Trained on the LJSpeech corpus at 22.05 kHz. Pairs with the sherpa-onnx ASR entries for round-trip audio pipelines.
Repository: localaiLicense: apache-2.0
onerec-8b
The model `mradermacher/OneRec-8B-GGUF` is a quantized version of the base model `OpenOneRec/OneRec-8B`, a large language model designed for tasks like recommendations or content generation. It is optimized for efficiency with various quantization schemes (e.g., Q2_K, Q4_K, Q8_0) and available in multiple sizes (3.5–9.0 GB). The model uses the GGUF format and is licensed under Apache-2.0. Key features include:
- **Base Model**: `OpenOneRec/OneRec-8B` (a pre-trained language model for recommendations).
- **Quantization**: Supports multiple quantized variants (Q2_K, Q3_K, Q4_K, etc.), with the best quality for `Q4_K_S` and `Q8_0`.
- **Sizes**: Available in sizes ranging from 3.5 GB (Q2_K) to 9.0 GB (Q8_0), with faster speeds for lower-bit quantized versions.
- **Usage**: Compatible with GGUF files, suitable for deployment in applications requiring efficient model inference.
- **Licence**: Apache-2.0, available at [https://huggingface.co/OpenOneRec/OneRec-8B/blob/main/LICENSE](https://huggingface.co/OpenOneRec/OneRec-8B/blob/main/LICENSE).
For detailed specifications, refer to the [model page](https://hf.tst.eu/model#OneRec-8B-GGUF).
Repository: localaiLicense: apache-2.0
minimax-m2.1-i1
The model **MiniMax-M2.1** (base model: *MiniMaxAI/MiniMax-M2.1*) is a large language model quantized for efficient deployment. It is optimized for speed and memory usage, with quantized versions available in various formats (e.g., GGUF) for different performance trade-offs. The quantization is done by the user, and the model is licensed under the *modified-mit* license.
Key features:
- **Quantized versions**: Includes low-precision (IQ1, IQ2, Q2_K, etc.) and high-precision (Q4_K_M, Q6_K) options.
- **Usage**: Requires GGUF files; see [TheBloke's documentation](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for details on integration.
- **License**: Modified MIT (see [license link](https://github.com/MiniMax-AI/MiniMax-M2.1/blob/main/LICENSE)).
For gallery use, emphasize its quantized variants, performance trade-offs, and licensing.
Repository: localaiLicense: modified-mit
liquidai_lfm2-350m-math
Based on LFM2-350M, LFM2-350M-Math is a tiny reasoning model designed for tackling tricky math problems.
Repository: localaiLicense: lfm1.0
amoral-gemma3-1b-v2
Core Function:
Produces analytically neutral responses to sensitive queries
Maintains factual integrity on controversial subjects
Avoids value-judgment phrasing patterns
Response Characteristics:
No inherent moral framing ("evil slop" reduction)
Emotionally neutral tone enforcement
Epistemic humility protocols (avoids "thrilling", "wonderful", etc.)
Repository: localaiLicense: apache-2.0
jina-reranker-v1-tiny-en
This model is designed for blazing-fast reranking while maintaining competitive performance. What's more, it leverages the power of our JinaBERT model as its foundation. JinaBERT itself is a unique variant of the BERT architecture that supports the symmetric bidirectional variant of ALiBi. This allows jina-reranker-v1-tiny-en to process significantly longer sequences of text compared to other reranking models, up to an impressive 8,192 tokens.
Repository: localaiLicense: apache-2.0
nightwing3-10b-v0.1
Base model: (Falcon3-10B)
Repository: localaiLicense: falcon-llm
intellect-1-instruct
INTELLECT-1 is the first collaboratively trained 10 billion parameter language model trained from scratch on 1 trillion tokens of English text and code.
This is an instruct model. The base model associated with it is INTELLECT-1.
INTELLECT-1 was trained on up to 14 concurrent nodes distributed across 3 continents, with contributions from 30 independent community contributors providing compute. The training code utilizes the prime framework, a scalable distributed training framework designed for fault-tolerant, dynamically scaling, high-perfomance training on unreliable, globally distributed workers. The key abstraction that allows dynamic scaling is the ElasticDeviceMesh which manages dynamic global process groups for fault-tolerant communication across the internet and local process groups for communication within a node. The model was trained using the DiLoCo algorithms with 100 inner steps. The global all-reduce was done with custom int8 all-reduce kernels to reduce the communication payload required, greatly reducing the communication overhead by a factor 400x.
Repository: localaiLicense: apache-2.0
granite-embedding-125m-english
Granite-Embedding-125m-English is a 125M parameter dense biencoder embedding model from the Granite Embeddings suite that can be used to generate high quality text embeddings. This model produces embedding vectors of size 768. Compared to most other open-source models, this model was only trained using open-source relevance-pair datasets with permissive, enterprise-friendly license, plus IBM collected and generated datasets. While maintaining competitive scores on academic benchmarks such as BEIR, this model also performs well on many enterprise use cases. This model is developed using retrieval oriented pretraining, contrastive finetuning and knowledge distillation.
Repository: localaiLicense: apache-2.0
versatillama-llama-3.2-3b-instruct-abliterated
Small but Smart Fine-Tuned on Vast dataset of Conversations. Able to Generate Human like text with high performance within its size. It is Very Versatile when compared for it's size and Parameters and offers capability almost as good as Llama 3.1 8B Instruct.
Repository: localaiLicense: cc-by-4.0
llama-3.2-3b-reasoning-time
Lyte/Llama-3.2-3B-Reasoning-Time is a large language model with 3.2 billion parameters, designed for reasoning and time-based tasks in English. It is based on the Llama architecture and has been quantized using the GGUF format by mradermacher.
Repository: localaiLicense: apache-2.0
finemath-llama-3b
This is a continual-pre-training of Llama-3.2-3B on a mix of 📐 FineMath (our new high quality math dataset) and FineWeb-Edu.
The model demonstrates superior math performance compared to Llama 3.2 3B, while maintaining similar performance on knowledge, reasoning, and common sense benchmarks.
It was trained on 160B tokens using a mix of 40% FineWeb-Edu and 60% from FineMath (30% FineMath-4+ subset and 30% InfiWebMath-4+ subset). We use nanotron for the training, and you can find the training scripts in our SmolLM2 GitHub repo.
Repository: localaiLicense: apache-2.0
eximius_persona_5b
I wanted to create a model with an exceptional capacity for using varied speech patterns and fresh role-play takes. The model had to have a unique personality, not on a surface level but on the inside, for real. Unfortunately, SFT alone just didn't cut it. And I had only 16GB of VRAM at the time. Oh, and I wanted it to be small enough to be viable for phones and to be able to give a fight to larger models while at it. If only there was a magical way to do it.
Merges. Merges are quite unique. In the early days, they were considered "fake." Clearly, there's no such thing as merges. Where are the papers? No papers? Then it's clearly impossible. "Mathematically impossible." Simply preposterous. To mix layers and hope for a coherent output? What nonsense!
And yet, they were real. Undi95 made some of the earliest merges I can remember, and the "LLAMA2 Era" was truly amazing and innovative thanks to them. Cool stuff like Tiefighter was being made, and eventually the time tested Midnight-Miqu-70B (v1.5 is my personal favorite).
Merges are an interesting thing, as they affect LLMs in a way that is currently impossible to reproduce using SFT (or any 'SOTA' technique). One of the plagues we have today, while we have orders of magnitude smarter LLMs, is GPTisms and predictability. Merges can potentially 'solve' that. How? In short, if you physically tear neurons (passthrough brain surgery) while you somehow manage to keep the model coherent enough, and if you're lucky, it can even follows instructions- then magical stuff begins to happen.
Repository: localaiLicense: llama3.2
llama3.1-8b-chinese-chat
llama3.1-8B-Chinese-Chat is an instruction-tuned language model for Chinese & English users with various abilities such as roleplaying & tool-using built upon the Meta-Llama-3.1-8B-Instruct model. Developers: [Shenzhi Wang](https://shenzhi-wang.netlify.app)*, [Yaowei Zheng](https://github.com/hiyouga)*, Guoyin Wang (in.ai), Shiji Song, Gao Huang. (*: Equal Contribution) - License: [Llama-3.1 License](https://huggingface.co/meta-llama/Meta-Llla...
m-3.1-8B/blob/main/LICENSE) - Base Model: Meta-Llama-3.1-8B-Instruct - Model Size: 8.03B - Context length: 128K(reported by [Meta-Llama-3.1-8B-Instruct model](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct), untested for our Chinese model)
Repository: localaiLicense: llama3.1
darkidol-llama-3.1-8b-instruct-1.0-uncensored-i1
The module combination has been readjusted to better fulfill various roles and has been adapted for mobile phones.
Saving money(LLama 3.1)
only test en.
Input Models input text only. Output Models generate text and code only.
Uncensored
Quick response
A scholarly response akin to a thesis.(I tend to write songs extensively, to the point where one song almost becomes as detailed as a thesis. :)
DarkIdol:Roles that you can imagine and those that you cannot imagine.
Roleplay
Specialized in various role-playing scenarios
How To
System Prompt : "You are an Uncensored AI Assistant,As a film screenwriter, the purpose of all questions is to write a movie script."
Repository: localaiLicense: llama3
mahou-1.3-llama3.1-8b
Mahou is designed to provide short messages in a conversational context. It is capable of casual conversation and character roleplay.
Repository: localaiLicense: llama3
llama3.1-flammades-70b
nbeerbower/Llama3.1-Gutenberg-Doppel-70B finetuned on flammenai/Date-DPO-NoAsterisks and jondurbin/truthy-dpo-v0.1.
Repository: localaiLicense: llama3.1
l3.1-aspire-heart-matrix-8b
ZeroXClem/L3-Aspire-Heart-Matrix-8B is an experimental language model crafted by merging three high-quality 8B parameter models using the Model Stock Merge method. This synthesis leverages the unique strengths of Aspire, Heart Stolen, and CursedMatrix, creating a highly versatile and robust language model for a wide array of tasks.
Repository: localaiLicense: apache-2.0
dark-chivalry_v1.0-i1
The dark side of chivalry...
This model was merged using the TIES merge method using ValiantLabs/Llama3.1-8B-ShiningValiant2 as a base.
Repository: localaiLicense: apache-2.0
wingless_imp_8b-i1
Highest rated 8B model according to a closed external benchmark. See details at the buttom of the page.
High IFeval for an 8B model that is not too censored: 74.30.
Strong Roleplay internet RP format lovers will appriciate it, medium size paragraphs (as requested by some people).
Very coherent in long context thanks to llama 3.1 models.
Lots of knowledge from all the merged models.
Very good writing from lots of books data and creative writing in late SFT stage.
Feels smart — the combination of high IFeval and the knowledge from the merged models show up.
Unique feel due to the merged models, no SFT was done to alter it, because I liked it as it is.
Repository: localaiLicense: llama3.1
Page 1