The Pikachu Language Model is a unique experimental language project initiated by Clemylia. It is trained completely from scratch and is based solely on the sound corpus of 'Pikachu', demonstrating the ability of small language models to create a native language identity.

Natural Language ProcessingTransformers

Gheya-1 is a new-generation basic language model in the LES-IA-ETOILES ecosystem, with 202 million parameters. It is an upgraded version of the old Small-lamina series. This model is designed specifically for professional fine-tuning and has targeted training in the fields of artificial intelligence, professional language models, and biology.

Natural Language ProcessingTransformersFrench

Fara-7B is a small language model developed by Microsoft Research, specifically designed for computer usage scenarios. It has only 7 billion parameters and achieves excellent performance among models of the same scale. It can perform computer interaction tasks such as web automation and multimodal understanding.

MultimodalTransformersEnglish

Granite 4.0 H-Small (FP8) is a small FP8 quantized version of the Granite series of language models developed by IBM, designed to provide professional, accurate, and secure responses. The model uses the FP8 precision format, optimizing inference efficiency while maintaining performance.

Natural Language ProcessingTransformers

trlm-135m is a small research prototype language model with 135 million parameters. It is built on SmolLM2-135M-Instruct and has undergone a three-stage fine-tuning process to explore how small language models can acquire step-by-step reasoning abilities.

Natural Language ProcessingTransformersEnglish

QuestA is an innovative framework that enhances the reasoning ability of large language models through a question-enhanced approach. It integrates partial solutions during the reinforcement learning training process, significantly improving the model's performance on complex tasks such as mathematical reasoning, especially achieving optimal results on small-parameter models.

Natural Language ProcessingTransformers

MobileLLM-R1 is a series of efficient inference language models released by Meta, focusing on solving mathematical, programming, and scientific problems. This model can still achieve excellent performance with a relatively small parameter scale, and provides a complete training recipe and data sources to support reproducible research.

Natural Language ProcessingTransformersEnglish

Cthulhu is a large model fusion project based on Mistral Small v3.2 and v3.1. It is a super powerful 24B parameter language model created by fusing multiple cutting-edge fine-tuned models. It has octopus-like multifaceted features, providing uncensored creative text generation capabilities, and excels in prose writing, instruction following, and esoteric knowledge expression.

Natural Language ProcessingTransformersEnglish

Cthulhu 24B 1.1 GGUF is a large language model fused based on the Mistral Small series of models, integrating 17 different model components. It has the capabilities of creative writing, uncensored dialogue, and rich language interaction. Inspired by the Cthulhu mythos, this model aims to provide a'super fusion' experience that goes beyond the capabilities of a single model.

Natural Language ProcessingTransformersEnglish

SmolLM3 is a language model with 3 billion parameters, aiming to break the performance boundary of small models. It supports 6 languages, has advanced reasoning ability and long context processing ability, and is a fully open-source model.

Natural Language ProcessingTransformersMultiple Languages

Qwen3-0.6B is the latest generation of small large language models in the Tongyi Qianwen series, with 0.6B parameters. It supports switching between thinking and non-thinking modes and is suitable for text generation, reasoning, and intelligent agent tasks.

Natural Language ProcessingTransformers

Ovis2-1B is the latest member of the Ovis series of multimodal large language models (MLLM), focusing on structural alignment of vision and text embeddings, featuring high performance for small models, enhanced reasoning capabilities, video and multi-image processing, and multilingual OCR enhancement.

MultimodalTransformersMultiple Languages

ThinkEdit is a lightweight weight editing method that identifies and edits a small number of attention heads to mitigate the issue of large language models generating overly short reasoning chains in inference tasks, thereby improving reasoning accuracy.

Natural Language ProcessingTransformers

Cuckoo is a small (300M parameters) information extraction (IE) model that mimics the next-token prediction paradigm of large language models by predicting the next token in a given context.

Natural Language ProcessingTransformers

Super Rainbow Cuckoo is a small-scale information extraction model based on the Next Token Extraction (NTE) paradigm, achieving efficient information extraction by mimicking the prediction methods of large language models.

Natural Language ProcessingTransformers

Cuckoo is a small (300M parameters) information extraction model that efficiently extracts information by mimicking the next-word prediction paradigm of large language models

Natural Language ProcessingTransformers

Atla Selene Mini is currently the most advanced small judge language model (SLMJ), with performance comparable to models 10 times its size, surpassing GPT-4o in multiple benchmarks.

Natural Language ProcessingTransformersMultiple Languages

This model is a fine-tuned LLM router classifier based on ModernBERT-large, used to distinguish between complex queries (using large language models) and simple queries (using small language models).

Natural Language ProcessingTransformers

TinyCodeLM is a series of small open - source code generation language models, offering two parameter scales of 150M and 400M. The model has undergone pre - training and instruction tuning, and performs excellently in Python code synthesis. It uses the Apache - 2.0 license.

Natural Language ProcessingTransformers

Mistral-Nemo-Instruct-2407 is a large language model fine-tuned on instructions based on Mistral-Nemo-Base-2407, jointly trained by Mistral AI and NVIDIA, outperforming existing models of similar or smaller size.

Natural Language ProcessingTransformersMultiple Languages