Join the conversation
Join the community of Machine Learners and AI enthusiasts.
Sign UpAll HF Hub posts
hypothetical
posted an update 2 days ago
Post
3581
The smallest and the highest quality in the world Gemma4 E2B and E4B models! 7x compression! From 9.3GB -> 1.4GB!
TheStageAI/gemma-4-E2B-it
TheStageAI/gemma-4-E4B-it
TheStageAI/gemma-4-E2B-it
TheStageAI/gemma-4-E4B-it
- 2 replies
danielhanchen
posted an update about 17 hours ago
Post
805
Gemma 4 12B can now run locally on just 8GB RAM via Dynamic GGUFs.
Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.
GGUF: unsloth/gemma-4-12b-it-GGUF
Guide: https://unsloth.ai/docs/models/gemma-4
Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.
GGUF: unsloth/gemma-4-12b-it-GGUF
Guide: https://unsloth.ai/docs/models/gemma-4
sergiopaniego
posted an update 3 days ago
Post
6167
new banger blog alert 🚨
@ariG23498 is starting a blog series about profiling in pytorch and part 1 just dropped
takes you from the simplest scenario to actually knowing what your gpu is doing. if you have never opened a profiler trace this is where you start
covers
find it here: https://huggingface.co/blog/torch-profiler
@ariG23498 is starting a blog series about profiling in pytorch and part 1 just dropped
takes you from the simplest scenario to actually knowing what your gpu is doing. if you have never opened a profiler trace this is where you start
covers
torch.profiler from scratch. reading tables and traces, overhead bound vs compute bound, the full dispatch chain from python to gpu kernels, and what torch.compile is actually fusing under the hoodfind it here: https://huggingface.co/blog/torch-profiler
- 1 reply
Post
3084
Hugging Face MCP Server v0.3.17
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SEP-2640 "Skills Over MCP" support added (early access)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SEP-2640 "Skills Over MCP" support added (early access)
- 1 reply
Post
728
New blog post!
An introduction to a little-known but highly effective model reduction method: 𝗧𝗿𝗶𝗺𝗺𝗶𝗻𝗴✂️
We show how to reduce model size (we went up to 87.24% reduction) while preserving its performance.
We applied this technique to 16 different model families across several modalities to illustrate that it works on any architecture (as long as the embedding layer is the last one of the model) and on any modality involving text.
From these 16 families, we generated over 𝟱,𝟱𝟬𝟬 𝗺𝗼𝗻𝗼𝗹𝗶𝗻𝗴𝘂𝗮𝗹 𝗺𝗼𝗱𝗲𝗹𝘀 𝗶𝗻 𝟭𝟮𝟰 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲𝘀 🌍
Key takeaways from our experiments:
1️⃣ Trimming does not require a GPU. Our models were obtained on a CPU.
2️⃣ This method scales up to at least 4B parameters (we did not test beyond that).
3️⃣ Trimmed model is smaller than the original while preserving its performance. If you observe a slight performance drop, just fine-tuned to recover or even surpass the original performance.
4️⃣ For an equivalent compute budget, it is better to trim then fine-tune rather than fine-tuning the original model. Since the model is smaller, you can run more epochs/show more data and get in fine a better model than the original.
5️⃣ Trimming is a competitive alternative to distillation and quantization. E.g. we obtained our alternative to DistilBERT in 9 minutes on CPU vs. 90 hours of GPU for the latter.
6️⃣ Trimming could generate reasoning traces in the language of the trimmed model. This could be an alternative to generating traces in English and then translating them into the desired language.
And many other things (such as how much data are needed, the impact of the database used, the order in which it should be done, etc.) are available in the blogpost!
Blogpost: https://huggingface.co/blog/lbourdois/introduction-to-trimming
Models: alphaedge-ai/Trimming_models_search
An introduction to a little-known but highly effective model reduction method: 𝗧𝗿𝗶𝗺𝗺𝗶𝗻𝗴✂️
We show how to reduce model size (we went up to 87.24% reduction) while preserving its performance.
We applied this technique to 16 different model families across several modalities to illustrate that it works on any architecture (as long as the embedding layer is the last one of the model) and on any modality involving text.
From these 16 families, we generated over 𝟱,𝟱𝟬𝟬 𝗺𝗼𝗻𝗼𝗹𝗶𝗻𝗴𝘂𝗮𝗹 𝗺𝗼𝗱𝗲𝗹𝘀 𝗶𝗻 𝟭𝟮𝟰 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲𝘀 🌍
Key takeaways from our experiments:
1️⃣ Trimming does not require a GPU. Our models were obtained on a CPU.
2️⃣ This method scales up to at least 4B parameters (we did not test beyond that).
3️⃣ Trimmed model is smaller than the original while preserving its performance. If you observe a slight performance drop, just fine-tuned to recover or even surpass the original performance.
4️⃣ For an equivalent compute budget, it is better to trim then fine-tune rather than fine-tuning the original model. Since the model is smaller, you can run more epochs/show more data and get in fine a better model than the original.
5️⃣ Trimming is a competitive alternative to distillation and quantization. E.g. we obtained our alternative to DistilBERT in 9 minutes on CPU vs. 90 hours of GPU for the latter.
6️⃣ Trimming could generate reasoning traces in the language of the trimmed model. This could be an alternative to generating traces in English and then translating them into the desired language.
And many other things (such as how much data are needed, the impact of the database used, the order in which it should be done, etc.) are available in the blogpost!
Blogpost: https://huggingface.co/blog/lbourdois/introduction-to-trimming
Models: alphaedge-ai/Trimming_models_search
- 4 replies
AxionLab-official
posted an update 1 day ago
Post
2334
We're happy to announce that we released a Reasoning tuned version of Supra-50M!
SupraLabs/Supra-50M-Reasoning
SupraLabs/Supra-50M-Reasoning
RiverRider
posted an update 2 days ago
Post
2715
This is not the end of words. It is the end of pretending their meanings are determined.
Meaning Forks. SRT detects it.
Paste any text to identify contested terms
RiverRider/srt-introspect
Try any prompt (attached link) to see exactly what an LLM is thinking at every meaningful step of its answer
RiverRider/srt-introspect
Repository
https://github.com/space-bacon/SRT
Paper
https://github.com/space-bacon/SRT/blob/main/paper_nla.md
Explainer
https://github.com/space-bacon/SRT/blob/main/docs/EXPLAINERS.md
Meaning Forks. SRT detects it.
Paste any text to identify contested terms
RiverRider/srt-introspect
Try any prompt (attached link) to see exactly what an LLM is thinking at every meaningful step of its answer
RiverRider/srt-introspect
Repository
https://github.com/space-bacon/SRT
Paper
https://github.com/space-bacon/SRT/blob/main/paper_nla.md
Explainer
https://github.com/space-bacon/SRT/blob/main/docs/EXPLAINERS.md
Post
2481
Qwen Image Edit 2511 Fast + LoRA ⚡
ovi054/Qwen-Image-Edit-2511-LoRA
QIE-2511 is an image editing model with integrated LoRA capabilities. You can add any custom LoRA to generate and edit images within this Space.
👉 Try it now: ovi054/Qwen-Image-Edit-2511-LoRA
ovi054/Qwen-Image-Edit-2511-LoRA
QIE-2511 is an image editing model with integrated LoRA capabilities. You can add any custom LoRA to generate and edit images within this Space.
👉 Try it now: ovi054/Qwen-Image-Edit-2511-LoRA
sergiopaniego
posted an update 2 days ago
Post
2233
most multi-turn RL loops have a silent bug: you decode the model's output to detect tool calls, then re-tokenize the conversation for the next turn. BPE isn't invertible, so decode then re-encode can land on different ids. gradient ends up on tokens the model never sampled. no crash, just quietly wrong math and broken training
@qgallouedec wrote a super educational blog on MITO (message-in, token-out) vs TITO (token-in, token-out) and how you might fix the problem above
go read it 🤓
https://qgallouedec-tito.hf.space/
@qgallouedec wrote a super educational blog on MITO (message-in, token-out) vs TITO (token-in, token-out) and how you might fix the problem above
go read it 🤓
https://qgallouedec-tito.hf.space/
