全站搜索
AI 模型 84759
- encoder↗
chatpig
- Llama3-8B-Chinese-Chat-ExPO↗
chujiezheng
The extrapolated (ExPO) model based on `shenzhi-wang/Llama3-8B-Chinese-Chat` and `meta-llama/Meta-Llama-3-8B-Instruct`, as in the "Weak-to-Strong Extrapolation Expedites Alignment" paper. Specifically, we obtain this model by extrapolating **(alpha = 0.3)** from the weights of the SFT and DPO/RLHF checkpoints, achieving superior alignment with human preference. **Note:** This is an experimental model, as I have not comprehensively evaluated its Chinese ability. **Unexpected issues may occur when we apply extrapolation to the DPO/RLHF alignment training for new languages (e.g., Chinese).**
- MiniMax-Music3-GGUF↗
ckadirt
- ContextPilot-14B↗
腾讯
- semsimula-ladder-live-owt-d384-l4-none↗
dimitarpg13
- wav2vec2-emotion-recognition↗
Dpngtm
This model is a fine-tuned version of facebook/wav2vec2-base-960h for **Speech Emotion Recognition (SER)**. It has been trained using a **Frozen Feature Extractor** strategy to preserve the model's acoustic understanding while adapting to emotion detection. This approach ensures stable performance and prevents "Catastrophic Forgetting," achieving nearly **80% accuracy** on the validation set. **Update:** The "Calm" and "Neutral" classes have been merged to improve classification consistency, resulting in **7 distinct emotion classes**.
- flux-3-image-klein↗
drawthingsai
FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing. It takes a prompt and up to 10 reference images and returns one image. Text-to-image, single-reference edits and multi-reference compositions run through the same model, with Qwen3-VL-4B-Instruct as the text encoder. This repository holds the distilled checkpoint for generation and image editing. For fine-tuning, LoRA training and research, use FLUX 3 Image Klein Base. Inference code, prompt upsampling and the bounding box format: black-forest-labs/flux-3-image-klein on GitHub.
- Qwen-Edit-2509-Multi-Angle-Lighting↗
dx8152
1. Overall performance was better 2. The brightness of the light can be controlled using the black and white brightness settings shown in Figure 2. 3. The lighting at some angles is still out of control; this will be fixed in the next update. 默认提示词：“使用图2的亮度贴图对图1重新照明(光源来自前方)” 只修改括号内的光源方向即可，例如现在的提示词是“光源来自前方”，可以修改为： 前方、左前方、左方、左后方、后方、右后方、右方、右前方、上方、下方 The default prompt message is: "Relight Figure 1 using the luminance map from Figure 2 (light source from the front)". You only need to modify the light source direction within the brackets. For example, the current prompt message is "light source from the front", which can be changed to: Front, Left Front, Left, Left Rear, Rear, Right Rear, Right, Right Front, Above, Below
- pythia-31m-deduped↗
EleutherAI
- gliner_medium_news-v2.1↗
EmergentMethods
This model is a fine-tune of GLiNER aimed at improving accuracy across a broad range of topics, especially with respect to long-context news entity extraction. As shown in the table below, these fine-tunes improved upon the base GLiNER model zero-shot accuracy by up to 7.5% across 18 benchmark datasets. The underlying dataset, AskNews-NER-v0 was engineered with the objective of diversifying global perspectives by enforcing country/language/topic/temporal diversity. All data used to fine-tune this model was synthetically generated. WizardLM 13B v1.2 was used for translation/summarization of open-web news articles, while Llama3 70b instruct was used for entity extraction. Both the diversification and fine-tuning methods are presented in a our paper on ArXiv. The synthetic data underlying this news fine-tune was pulled from the AskNews API. We enforced diveristy across country/language/topic/time.