8
Lightweight Embeddings API
π»
Generate embeddings and rerank text or images
Hieu Lam
lamhieu
AI & ML interests
.-.
Recent Activity
updated
a Space
about 7 hours ago
lamhieu/lightweight-embeddings
liked
a Space
4 days ago
baohuynhbk14/Qwen3-VL-Demo
replied to
their
post
7 days ago
π Introducing the xLLMs Dataset Collection
The xLLMs project is a growing suite of multilingual and multimodal dialogue datasets designed to train and evaluate advanced conversational LLMs. Each dataset focuses on a specific capability β from long-context reasoning and factual grounding to STEM explanations, math Q&A, and polite multilingual interaction.
π Explore the full collection on Hugging Face:
π https://huggingface.co/collections/lamhieu/xllms-66cdfe34307bb2edc8c6df7d
π¬ Highlight: xLLMs β Dialogue Pubs
A large-scale multilingual dataset built from document-guided synthetic dialogues (Wikipedia, WikiHow, and technical sources). Itβs ideal for training models on long-context reasoning, multi-turn coherence, and tool-augmented dialogue across 9 languages.
π https://huggingface.co/datasets/lamhieu/xllms_dialogue_pubs
π§ Designed for:
- Long-context and reasoning models
- Multilingual assistants
- Tool-calling and structured response learning
All datasets are open for research and development use β free, transparent, and carefully curated to improve dialogue model quality.
Organizations
updated
a
Space
about 7 hours ago
liked
a
Space
4 days ago
replied to
their
post
7 days ago
using to build tf great ;)
liked
a
model
7 days ago
liked
a
model
8 days ago
replied to
their
post
8 days ago
go ahead and create something interesting ;)
posted
an
update
8 days ago
Post
2679
π Introducing the xLLMs Dataset Collection
The xLLMs project is a growing suite of multilingual and multimodal dialogue datasets designed to train and evaluate advanced conversational LLMs. Each dataset focuses on a specific capability β from long-context reasoning and factual grounding to STEM explanations, math Q&A, and polite multilingual interaction.
π Explore the full collection on Hugging Face:
π lamhieu/xllms-66cdfe34307bb2edc8c6df7d
π¬ Highlight: xLLMs β Dialogue Pubs
A large-scale multilingual dataset built from document-guided synthetic dialogues (Wikipedia, WikiHow, and technical sources). Itβs ideal for training models on long-context reasoning, multi-turn coherence, and tool-augmented dialogue across 9 languages.
π lamhieu/xllms_dialogue_pubs
π§ Designed for:
- Long-context and reasoning models
- Multilingual assistants
- Tool-calling and structured response learning
All datasets are open for research and development use β free, transparent, and carefully curated to improve dialogue model quality.
The xLLMs project is a growing suite of multilingual and multimodal dialogue datasets designed to train and evaluate advanced conversational LLMs. Each dataset focuses on a specific capability β from long-context reasoning and factual grounding to STEM explanations, math Q&A, and polite multilingual interaction.
π Explore the full collection on Hugging Face:
π lamhieu/xllms-66cdfe34307bb2edc8c6df7d
π¬ Highlight: xLLMs β Dialogue Pubs
A large-scale multilingual dataset built from document-guided synthetic dialogues (Wikipedia, WikiHow, and technical sources). Itβs ideal for training models on long-context reasoning, multi-turn coherence, and tool-augmented dialogue across 9 languages.
π lamhieu/xllms_dialogue_pubs
π§ Designed for:
- Long-context and reasoning models
- Multilingual assistants
- Tool-calling and structured response learning
All datasets are open for research and development use β free, transparent, and carefully curated to improve dialogue model quality.
-
- 4 replies
updated
2
datasets
8 days ago
published
5
datasets
8 days ago
lamhieu/xllms_dialogue_pubs
Viewer
β’
Updated
β’
999k
β’
85
β’
4
lamhieu/xllms_dialogue_greetings
Viewer
β’
Updated
β’
41.3k
β’
17
β’
1
lamhieu/xllms_dialogue_wildchat
Viewer
β’
Updated
β’
206k
β’
21
β’
1
lamhieu/xllms_dialogue_stem
Viewer
β’
Updated
β’
110k
β’
14
lamhieu/xllms_dialogue_mathqa
Viewer
β’
Updated
β’
395k
β’
13
upvoted
a
collection
17 days ago
upvoted
an
article
17 days ago
Article
Swift Transformers Reaches 1.0 β and Looks to the Future
β’
34
upvoted
an
article
27 days ago
Article
`LeRobotDataset`: Bringing large-scale datasets to lerobot
β’
43
liked
a
model
2 months ago
replied to
their
post
2 months ago
At the company where I work, we always use paid tools to build features and support tools such as Copilot, OpenAI APIs, Gemini, Vertex, and so on. But for example, with Copilot, itβs very difficult to know whether they donβt use enterprise data for training. Of course, we always see those βtermsβ stated, but you know, the data passes through so many layers.
posted
an
update
2 months ago
Post
468
Copilot, Cursor and all those new AI-driven extensions are rising fast β they can read your codebase, optimize logic, and even automate whole workflows. But hereβs the flip side: once these tools start peeking into .env files, nodes, and source code, the risk of leaking critical keys or sensitive info becomes very real.
So yeah, weβre coding faster than ever, but are we also opening the door to a whole new wave of vulnerabilities?
π Is this the hidden threat waiting to hit the next era of βvibe-codingβ?
So yeah, weβre coding faster than ever, but are we also opening the door to a whole new wave of vulnerabilities?
π Is this the hidden threat waiting to hit the next era of βvibe-codingβ?
-
- 2 replies