FastVLM: Efficient Vision Encoding for Vision Language Models
Paper
•
2412.13303
•
Published
•
73
Daniel Bourke PRO
mrdbourke
AI & ML interests
Computer vision. Small on-device models. VLMs. High-quality tutorials.
Recent Activity
upvoted
a
paper
about 16 hours ago
FastVLM: Efficient Vision Encoding for Vision Language Models
reacted
to
IlyasMoutawwakil's
post
with 🔥
1 day ago
After 2 months of refinement, I'm happy to announce that a lot of Transformers' modeling code is now significantly more torch-compile & export-friendly 🔥
Why it had to be done 👇
PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph !
Transformers models are now easier to:
⚙️ Compile end-to-end with torch.compile backends
📦 Export reliably via torch.export and torch.onnx.export
🚀 Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes.
This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators.
We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere.
There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs.
PR in the comments ! More updates coming coming soon !
reacted
to
danielhanchen's
post
with ❤️
1 day ago
You can now fine-tune embedding models in our free Unsloth notebook! 🤗
Fine-tuning embedding models improves retrieval & RAG by aligning vectors to your domain-specific notion of similarity, improving search, clustering, and recommendations on your data.
⭐ Blog + Notebooks: https://unsloth.ai/docs/new/embedding-finetuning
Unsloth trains embedding models 1.8-3.3x faster with 20% less VRAM, 2x longer context & no accuracy loss vs. FA2 setups.
We'd like to thank Hugging Face and Unsloth contributor: electroglyph for making this possible!
Organizations
None yet
upvoted
a
paper
about 16 hours ago
reacted to
IlyasMoutawwakil's
post with 🔥
1 day ago
Post
2107
After 2 months of refinement, I'm happy to announce that a lot of Transformers' modeling code is now significantly more torch-compile & export-friendly 🔥
Why it had to be done 👇
PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph !
Transformers models are now easier to:
⚙️ Compile end-to-end with torch.compile backends
📦 Export reliably via torch.export and torch.onnx.export
🚀 Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes.
This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators.
We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere.
There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs.
PR in the comments ! More updates coming coming soon !
Why it had to be done 👇
PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph !
Transformers models are now easier to:
⚙️ Compile end-to-end with torch.compile backends
📦 Export reliably via torch.export and torch.onnx.export
🚀 Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes.
This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators.
We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere.
There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs.
PR in the comments ! More updates coming coming soon !
- 1 reply
reacted to
danielhanchen's
post with ❤️
1 day ago
Post
2357
You can now fine-tune embedding models in our free Unsloth notebook! 🤗
Fine-tuning embedding models improves retrieval & RAG by aligning vectors to your domain-specific notion of similarity, improving search, clustering, and recommendations on your data.
⭐ Blog + Notebooks: https://unsloth.ai/docs/new/embedding-finetuning
Unsloth trains embedding models 1.8-3.3x faster with 20% less VRAM, 2x longer context & no accuracy loss vs. FA2 setups.
We'd like to thank Hugging Face and Unsloth contributor: electroglyph for making this possible!
Fine-tuning embedding models improves retrieval & RAG by aligning vectors to your domain-specific notion of similarity, improving search, clustering, and recommendations on your data.
⭐ Blog + Notebooks: https://unsloth.ai/docs/new/embedding-finetuning
Unsloth trains embedding models 1.8-3.3x faster with 20% less VRAM, 2x longer context & no accuracy loss vs. FA2 setups.
We'd like to thank Hugging Face and Unsloth contributor: electroglyph for making this possible!
- 1 reply
liked
a
model
2 days ago
liked
2
models
3 days ago
upvoted
a
collection
3 days ago
liked
a
Space
3 days ago
upvoted
a
collection
3 days ago
updated
a
Space
4 days ago
liked
a
Space
4 days ago
upvoted
a
paper
4 days ago
liked
2
models
4 days ago
published
a
Space
4 days ago
reacted to
danielhanchen's
post with 🔥
4 days ago
Post
2486
Run GLM-4.7-Flash locally on your device with 24GB RAM!🔥
It's the best performing 30B model on SWE-Bench and GPQA. With 200K context, it excels at coding, agents, chat & reasoning.
GGUF: unsloth/GLM-4.7-Flash-GGUF
Guide: https://unsloth.ai/docs/models/glm-4.7-flash
It's the best performing 30B model on SWE-Bench and GPQA. With 200K context, it excels at coding, agents, chat & reasoning.
GGUF: unsloth/GLM-4.7-Flash-GGUF
Guide: https://unsloth.ai/docs/models/glm-4.7-flash
reacted to
mitkox's
post with 🚀
4 days ago
Post
1413
GLM-4.7-Flash is fast, good and cheap.
3,074 tokens/sec peak at 200k tokens context window on my desktop PC.
Works with Claude Code and opencode for hours. No errors, drop-in replacement of the Anthropic cloud AI.
MIT licensed, open weights, free for commercial use and modifications.
Supports speculative decoding using MTP, which is highly effective in mitigating latency.
Great for on device AI coding as AWQ 4bit at 18.5 GB. Hybrid inference on a single consumer GPU + CPU RAM.
3,074 tokens/sec peak at 200k tokens context window on my desktop PC.
Works with Claude Code and opencode for hours. No errors, drop-in replacement of the Anthropic cloud AI.
MIT licensed, open weights, free for commercial use and modifications.
Supports speculative decoding using MTP, which is highly effective in mitigating latency.
Great for on device AI coding as AWQ 4bit at 18.5 GB. Hybrid inference on a single consumer GPU + CPU RAM.
-
- 3 replies
Use default attention implementation with option to override
1
#2 opened 17 days ago
by
nvidia-oliver-holworthy
Add SDPA attention support to models
#3 opened 4 days ago
by
mrdbourke
liked
a
model
4 days ago
