overview for pepperfree

8

Infinity-Parser2 - Multimodal Document Parser (huggingface.co)

submitted 3 months ago by pepperfree@sh.itjust.works to c/localllama@sh.itjust.works

0 comments fedilink

Based on Qwen3.5 MoE 35B and Qwen3.5 2B

Copied from readme.

We are excited to release Infinity-Parser2, our latest flagship document understanding model. We offer two distinct variants to address diverse deployment constraints: Infinity-Parser2-Pro, optimized for maximum accuracy in precision-critical tasks, achieves state-of-the-art results on olmOCR-Bench (87.6%) and ParseBench (74.3%), surpassing frontier models including DeepSeek-OCR-2, PaddleOCR-VL-1.5, and MinerU-2.5. Infinity-Parser2-Flash, engineered for low-latency inference, delivers a 3.68x speedup over our previous Infinity-Parser-7B model. With significant upgrades to both our data engine and multi-task reinforcement learning approach, the model consolidates robust multi-modal parsing capabilities into a unified architecture, unlocking brand-new zero-shot capabilities across a wide range of real-world business scenarios.

DeepSeek dropped the V3.1 Weight by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

I wonder if we can extend the context length. It already fine-tuned with YaRN so we can't get free extend with that method.

DeepSeek dropped the V3.1 Weight by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 2 points 1 year ago

Everybody been rumoring about R2. So releasing this thing kinda unexpected

45

DeepSeek dropped the V3.1 Weight (huggingface.co)

submitted 1 year ago by pepperfree@sh.itjust.works to c/localllama@sh.itjust.works

19 comments fedilink

Not what we expected...

Fine tuned models for summarisation? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

So something like

Previously the text talk about [last summary]
[The instruction prompt]...
[Current chunk/paragraphs]

When DeepSeek V4 and R2? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

The RL is so good grok changed it's personality by changing small part of it's system prompt

When DeepSeek V4 and R2? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 3 points 1 year ago

When DeepSeek V4 and R2? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 3 points 1 year ago

They got the whole Twitter database. It's kinda the same with Gemini. But somehow Meta isn't catching up, maybe their llama 4 architecture isn't that stable to train.

When DeepSeek V4 and R2? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

There is new project which they share fine-tuned modernbert on some task. Here is the org https://huggingface.co/adaptive-classifier

When DeepSeek V4 and R2? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

It changed after Grok 3

187

When DeepSeek V4 and R2? (lemmy.nz)

submitted 1 year ago by pepperfree@sh.itjust.works to c/localllama@sh.itjust.works

26 comments fedilink

So image generation is where it's at? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 3 points 1 year ago

Lots of developer choose to write in CUDA as ROCm support back then is a mess.

So image generation is where it's at? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 4 points 1 year ago* (last edited 1 year ago)

No, you can run sd, flux based model inside the koboldcpp. You can try it out using the original koboldcpp in google colab. It loads gguf model. Related discussion on Reddit: https://www.reddit.com/r/StableDiffusion/comments/1gsdygl/koboldcpp_now_supports_generating_images_locally/

Edit: Sorry, I kinda missed the point, maybe I'm sleepy when writing that comment. Yeah, I agree that LLM need big memory to run which is one of it's downside. I remember someone doing comparison that API with token based pricing is cheaper that to run it locally. But, running image generation locally is cheaper than API with step+megapixel pricing.

MindLink-32B and MindLink-72B available on Huggingface by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 1 points 1 year ago

Skywork downfall

So image generation is where it's at? by pepperfree in c/localllama@sh.itjust.works

[-] pepperfree@sh.itjust.works 3 points 1 year ago

There is koboldcp-rocm fork. Koboldcpp itself has basic image generation. https://github.com/YellowRoseCx/koboldcpp-rocm

20

MindLink-32B and MindLink-72B available on Huggingface (sh.itjust.works)

submitted 1 year ago by pepperfree@sh.itjust.works to c/localllama@sh.itjust.works

2 comments fedilink

Built on Qwen, these models incorporate our latest advances in post-training techniques. MindLink demonstrates strong performance across various common benchmarks and is widely applicable in diverse AI scenarios.

72B 32B