AI & ML interests

None defined yet.

NymboΒ 
posted an update 10 days ago
view post
Post
4624
πŸš€ I've just shipped a major update to the Nymbo/Tools MCP server: the Agent_Terminal, a single "master tool" that cuts token usage by over 90%!

Anthropic found 98.7% context savings using code execution with MCP, Cloudflare published similar findings. This is my open-source implementation of the same idea.

# The Problem

Traditional MCP exposes every tool definition directly to the model. With 12 tools, that's thousands of tokens consumed *before the conversation even starts*. Each tool call also passes intermediate results through the context window β€” a 10,000-row spreadsheet? That's all going into context just to sum a column.

# The Solution: One Tool to Rule Them All

Agent_Terminal wraps all 12 tools (Web_Search, Web_Fetch, File_System, Generate_Image, Generate_Speech, Generate_Video, Deep_Research, Memory_Manager, Obsidian_Vault, Shell_Command, Code_Interpreter) into a single Python code execution gateway.

Instead of the model making individual tool calls, it writes Python code that orchestrates the tools directly:

# Search for Bitcoin price
result = Web_Search("current price of bitcoin", max_results=3)
print(result)


Don't know what tools are available? The agent can discover them at runtime:

print(search_tools('image'))  # Find tools by keyword
print(usage('Generate_Image'))  # Get full docs for a specific tool


The individual direct tool calls are all still there, but they can be disabled if using the Agent_Terminal. Try it now - https://www.nymbo.net/nymbot
  • 1 reply
Β·
lunarfluΒ 
posted an update 28 days ago
lunarfluΒ 
posted an update 28 days ago
view post
Post
482
The new King πŸ‘‘has arrived!

Moonshot AI now the top model on Hugging Face πŸ”₯
moonshotai/Kimi-K2-Thinking
lunarfluΒ 
posted an update 28 days ago
view post
Post
2659
πŸ’ΈπŸ€‘You don’t need 100 GPUs to train something amazing!

Our Smol Training Playbook teaches you a better path to world-class LLMs, for free!

Check out the #1 trending space on πŸ€— :
HuggingFaceTB/smol-training-playbook
NymboΒ 
posted an update about 1 month ago
view post
Post
889
I've added an 11th tool to the Nymbo/Tools MCP server, it's for your Obsidian_Vault. I'd argue it's far more context-efficient than any other Obsidian MCP I've seen, and doesn't require any plugins. Also some big improvements to the Web_Search and Web_Fetch tools.

# Obsidian_Vault Tool

It's basically a read-only version of the File_System tool, but it works so well for navigating Obsidian without unnecessary context. It supports recursive (full-text) search across the entire vault, and supports offset so the agent can "scroll" through a document without re-consuming tokens.

Run the server locally and set the OBSIDIAN_VAULT_ROOT environment variable to your vault's root path. If you don't use Obsidian, this is perfectly usable as simply a read-only filesystem.

# Web_Search Improvements

The Web_Search tool previously just used DuckDuckGo as a backend search engine, but now it also supports Bing, Brave, Yahoo, and Wikipedia. Default engine is auto which provides results from all backends in recommended order. Still doesn't require any kind of API or auth for Web_Search.

There's also a new date filter to limit results to those created in the past day, week, month, or year. Oh, and uhh, SafeSearch is now off by default :)

# Web_Fetch Improvements

As context-efficient as the Markdown mode is for web browsing, sometimes it does lose important context in the conversion from HTML to Markdown. So I've added a new HTML mode to the Web_Fetch tool that basically executes a cURL request on the URL, returning the full HTML page if necessary.

# A Note on Claude Skills

I've been having fun with the new File_System and Shell_Command tools. Using Claude Skills doesn't currently work in the public HF space because of environment restrictions, but using Skills works perfectly well running locally.

Happy building ~
NymboΒ 
posted an update about 2 months ago
view post
Post
1965
Two new tools added to the Nymbo/Tools MCP server, File_System and Shell_Exec. You can theoretically do basically anything with these two tools, and it should enable support for many Claude Skills.

GPT-5-Codex proves that for many cases, shell commands really are all you need, and Claude Skills seem to lean into this. The thing is, nothing about the design of Claude Skills actually restricts them to proprietary models!

# File_System

There's a new directory inside the repo called Filesystem, that's the agent's "root". It can perform the following actions : list, read, write, append, mkdir, move, copy, delete, info, help. It's able to keep this all within the scope of one tool call by making the Action field required and all other fields optional. Using a filesystem shouldn't require 15 different tools.

Files created in the public HF space live in the space's running container, and gets cleared when the space is restarted. When running the server locally, files are actually stored on disk.

# Shell_Exec

What good is a filesystem if you can't execute commands in that filesystem? This tool automatically detects if the server is running on Windows or Linux, and suggests using the appropriate shell (PowerShell/Bash). Both of these new tools require that the agent uses relative paths, rather than absolute paths. I could be convinced to back pedal on this.

# Closing Thoughts

The File_System and Shell_Exec tools aren't super polished yet, I'll continue to improve the agent's instructions and UX of using the new tools. Most of my testing was done with gpt-oss-20b and if it messes up, it gets the gist after one failed tool call. It should work perfectly fine for the GPU poor.
  • 1 reply
Β·
NymboΒ 
posted an update about 2 months ago
view post
Post
1836
I've made some improvements to my custom Deep_Research tool in the Nymbo/Tools MCP server. I've added a second LLM process and it still takes less than 1 minute to complete!

The original version of my Deep_Research tool would basically dump up to 50 fetched webpages onto the Researcher model (Qwen3-235B), with only a little bit of context shown from each page.

# New "Filterer" Process

The new process includes another LLM call before the researcher process. The Filterer (also Qwen3-235B) gets the query summary and the original 50 pages with low context, and decides which pages are most relevant to the research topic. The Filterer then outputs the URLs to the relevant pages, which are then re-fetched (with more context) and sent to the Researcher.

# Researcher Context

The Researcher now gets only the relevant webpages, then begins writing the report. When testing with 50 initial results, the researcher would often end up with 10-20 results of relevant context.

It still takes less than a minute to accomplish everything, thanks entirely to Cerebras inference. It now takes about 35-45 seconds to complete once the tool is run.

It's also worth noting that both the Filterer and Researcher now are provided the current time/date before they see the content, reducing hallucinations caused by knowledge cutoffs.
lunarfluΒ 
posted an update 2 months ago
view post
Post
2268
Cool stuff these past weeks on huggingface! πŸ€— πŸš€ !
β€’ πŸ“ˆTrackio, local-first W&B alternative
https://github.com/gradio-app/trackio/issues
β€’ 🌍EmbeddingGemma, 300M-param, multilingual embeddings, on-device
https://huggingface.co/blog/embeddinggemma
β€’ πŸ’»Open LLMs in VS Code (Inference Providers)
https://x.com/reach_vb/status/1966185427582497171
β€’ πŸ€–Smol2Operator GUI agents
https://huggingface.co/blog/smol2operator
β€’ πŸ–ΌοΈGradio visible watermarking
https://huggingface.co/blog/watermarking-with-gradio
NymboΒ 
posted an update 2 months ago
view post
Post
674
I have a few Sora-2 invites - 15509N
  • 1 reply
Β·
NymboΒ 
posted an update 3 months ago
view post
Post
1101
There's now a custom Deep_Research tool in my Nymbo/Tools MCP server! TL;DR: The agent using the tools writes a summary of your requests and up to five DuckDuckGo searches (up to 50 results). Each of the webpages found in the searches are then fetched and given to our researcher (Qwen3-235B-A22B-Thinking-2507). The researcher sees the summary, searched queries, and fetched links, then writes a thorough research report. The agent using the tool provides the user with a summary of the report and a link to download research_report.txt. The researcher's instructions are similar to some leaked Perplexity sys prompts.

# Deep_Research Tool

It accomplishes everything in under a minute so it doesn't hit MCP's 60 second timeout, mostly thanks to Cerebras. The only thing required to make this work is a HF_READ_TOKEN for inference.

The Deep_Research tool could certainly be improved. It still needs some sort of mechanism for sorting URLs based on importance (I've got some ideas but I don't want it to be the responsibility of the agent using the tool). I'll probably add a second researcher to filter out the bad sources before inferencing the big researcher. I'm hellbent on keeping this all within the scope of one tool call.

# More Fetch/Web Search Improvements

The Search_DuckDuckGo tool has been further enhanced. It now allows the agent to browse through all pages of results. The results also now include published date (if detected). It also now supports every DDG search types! Default DDG search is called text, but it can also now search by news, images, videos, and books.

The Fetch_Webpage tool now specifies how much of the page has been truncated, and cursor index, allowing it to pickup where it left off without re-consuming tokens. The model can now also choose to strip CSS selectors to remove excess noise, and there's a new URL Scraper mode that only returns URLs found on the full page.

More to come soon ~
AbhaykoulΒ 
posted an update 3 months ago
view post
Post
3099
πŸš€ Ever dreamed of training your own Large Language Model from scratch? What if I told you it doesn't require a supercomputer or PhD in ML? 🀯

Introducing LLM Trainer - the educational framework that makes LLM training accessible to EVERYONE! Whether you're on a CPU-only laptop or scaling to distributed GPUs, we've got you covered. πŸ’»βž‘οΈπŸ–₯️

Why LLM Trainer? Because existing tools are either too simplistic (hiding the magic) or too complex (requiring expert knowledge). We bridge the gap with:

πŸŽ“ Educational transparency - every component built from scratch with clear code
πŸ’» CPU-first approach - start training immediately, no GPU needed
πŸ”§ Full customization - modify anything you want
πŸ“ˆ Seamless scaling - from laptop to cluster without code changes
🀝 HuggingFace integration - works with existing models & tokenizers

Key highlights:
βœ… Built-in tokenizers (BPE, WordPiece, HF wrappers)
βœ… Complete Transformer implementation from scratch
βœ… Optimized for CPU training
βœ… Advanced features: mixed precision, gradient checkpointing, multiple generation strategies
βœ… Comprehensive monitoring & metrics

Perfect for:
- Students learning transformers
- Researchers prototyping new ideas
- Developers building domain-specific models

Ready to train your first LLM? It's easier than you think!

πŸ”— Check it out: https://github.com/HelpingAI/llm-trainer
πŸ“š Docs: Getting Started Guide
πŸ’¬ Join the community: GitHub Discussions

#AI #MachineLearning #LLM #DeepLearning #OpenSource #Python #HuggingFace #NLP

Special thanks to HuggingFace and PyTorch teams for the amazing ecosystem! πŸ™
  • 1 reply
Β·
NymboΒ 
posted an update 3 months ago
view post
Post
1051
I have a few updates to my MCP server I wanna share: New Memory tool, improvements to web search & speech generation.

# Memory_Manager Tool

We now have a Memory_Manager tool. Ask ChatGPT to write all its memories verbatim, then tell gpt-oss-20b to save each one using the tool, then take them anywhere! It stores memories in a memories.json file in the repo, no external database required.

The Memory_Manager tool is currently hidden from the HF space because it's intended for local use. It's enabled by providing a HF_READ_TOKEN in the env secrets, although it doesn't actually use the key for anything. There's probably a cleaner way of ensuring memory is only used locally, I'll come back to this.

# Fetch & Websearch

The Fetch_Webpage tool has been simplified a lot. It now converts the page to Markdown and returns the page with three length settings (Brief, Standard, Full). This is a lot more reliable than the old custom extraction method.

The Search_DuckDuckGo tool has a few small improvements. The input is easier for small models to get right, and the output is more readable.

# Speech Generation

I've added the remaining voices for Kokoro-82M, it now supports all 54 voices with all accents/languages.

I also removed the 30 second cap by making sure it computes all chunks in sequence, not just the first. I've tested it on outputs that are ~10 minutes long. Do note that when used as an MCP server, the tool will timeout after 1 minute, nothing I can do about that for right now.

# Other Thoughts

Lots of MCP use cases involve manipulating media (image editing, ASR, etc.). I've avoided adding tools like this so far for two reasons:

1. Most of these solutions would require assigning it a ZeroGPU slot.
2. The current process of uploading files like images to a Gradio space is still a bit rough. It's doable but requires additional tools.

Both of these points make it a bit painful for local usage. I'm open to suggestions for other tools that rely on text.
NymboΒ 
posted an update 3 months ago
view post
Post
1029
I built a general use MCP space ~ Fetch webpages, DuckDuckGo search, Python code execution, Kokoro TTS, Image Gen, Video Gen.

# Tools

1. Fetch webpage
2. Web search via DuckDuckGo (very concise, low excess context)
3. Python code executor
4. Kokoro-82M speech generation
5. Image Generation (use any model from HF Inference Providers)
6. Video Generation (use any model from HF Inference Providers)

The first four tools can be used without any API keys whatsoever. DDG search is free and the code execution and speech gen is done on CPU. Having a HF_READ_TOKEN in the env variables will show all tools. If there isn't a key present, The Image/Video Gen tools are hidden.

Nymbo/Tools
  • 1 reply
Β·
NymboΒ 
posted an update 4 months ago
view post
Post
1047
Anyone using Jan-v1-4B for local MCP-based web search, I highly recommend you try out Intelligent-Internet/II-Search-4B

Very impressed with this lil guy and it deserves more downloads. It's based on the original version of Qwen3-4B but find that it questions reality way less often. Jan-v1 seems to think that everything it sees is synthetic data and constantly gaslights me
KingNishΒ 
posted an update 4 months ago
VokturzΒ 
posted an update 4 months ago
view post
Post
527
πŸ€— A Transformers.js Playground

I just created a new spaces to test models directly in the browser using Transformers.js. You can pick any model from a list (mainly from the onnx-community), or just load the one you desire :) It also allows you to choose among different quantizations!

Currently the following pipelines are supported
- feature-extraction
- image-classification
- text-generation
- text-classification
- text-to-speech
- zero-shot-classification

Vokturz/transformers-js-playground
AbhaykoulΒ 
posted an update 4 months ago
view post
Post
4132
πŸš€ Dhanishtha-2.0-preview-0825 Is Here

The Intermediate Thinking Model just leveled up again.

With sharper reasoning, better tool use, and expanded capabilities, Dhanishtha-2.0-preview-0825 is now live and ready to impress.

🧠 What Makes Dhanishtha Special?
Unlike typical CoT models that only thinks one time, Dhanishtha thinks iteratively:

> Think β†’ Answer β†’ Rethink β†’ Improve β†’ Rethink again if needed.

πŸ”— Try it now: HelpingAI/Dhanishtha-2.0-preview-0825

πŸ”ž Dhanishtha NSFW Preview

For those exploring more expressive and immersive roleplay scenarios, we’re also releasing:

HelpingAI/Dhanishtha-nsfw
A specialized version tuned for adult-themed interactions and character-driven roleplay.

πŸ”— Explore it here: HelpingAI/Dhanishtha-nsfw

πŸ’¬ You can also try all of these live at chat.helpingai.co
Β·
AtAndDevΒ 
posted an update 5 months ago
view post
Post
562
Qwen 3 Coder is a personal attack to k2, and I love it.
It achieves near SOTA on LCB while not having reasoning.
Finally people are understanding that reasoning isnt necessary for high benches...

Qwen ftw!

DECENTRALIZE DECENTRALIZE DECENTRALIZE
AbhaykoulΒ 
posted an update 5 months ago
view post
Post
3158
πŸŽ‰ Dhanishtha-2.0-preview-0725 is Now Live

The Intermediate Thinking Model just got even better.
With the new update, Dhanishtha is now sharper, smarter, and trained further on tool use

🧠 What Makes Dhanishtha Different?
Unlike standard COT models that give one-shot responses, Dhanishtha thinks in layers:

> Think β†’ Answer β†’ Rethink β†’ Improve β†’ Rethink again if needed.

HelpingAI/Dhanishtha-2.0-preview-0725
AbhaykoulΒ 
posted an update 5 months ago
view post
Post
3112
πŸŽ‰ Dhanishtha 2.0 Preview is Now Open Source!

The world's first Intermediate Thinking Model is now available to everyone!

Dhanishtha 2.0 Preview brings revolutionary intermediate thinking capabilities to the open-source community. Unlike traditional reasoning models that think once, Dhanishtha can think, answer, rethink, answer again, and continue rethinking as needed using multiple blocks between responses.

πŸš€ Key Features
- Intermediate thinking: Think β†’ Answer β†’ Rethink β†’ Answer β†’ Rethink if needed...
- Token efficient: Uses up to 79% fewer tokens than DeepSeek R1 on similar queries
- Transparent thinking: See the model's reasoning process in real-time
- Open source: Freely available for research and development


HelpingAI/Dhanishtha-2.0-preview
https://helpingai.co/chat
  • 1 reply
Β·