Hugging Face's logo
Activity Feed Request to join this org

AI & ML interests

Building interactive demos to scikit-learn examples ๐Ÿงก

Recent Activity
Q-bertย  authored a paper about 1 month ago
ARCHead: Activation-Metric Residual Correction for Large Language Model Output Heads
MElHuseyniย  authored a paper 5 months ago
RDP LoRA: Geometry-Driven Identification for Parameter-Efficient Adaptation in Large Language Models
MElHuseyniย  submitted a paper 5 months ago
RDP LoRA: Geometry-Driven Identification for Parameter-Efficient Adaptation in Large Language Models
View all activity
Aurelien-Morganย 
posted an update 7 days ago
view post
Post
2412
@retrain-pipelines execution engine is in perpetual evolution, with the aim to establish itself as SOTA, and for the long run.

However, we neglect no aspect of ML-Eng centricity.

If notebooks is where you like to do dev most,
we support you there 100% too.

Build crazy combos of inline tasks, deep parallel sub-DAG branches, nested asynchronous groups...

... the DAG renderer is undergoing an incremental upgrade

until the next one.

* starring toy tasks here. No ML has been hurt in this video ๐Ÿ™‚
AtAndDevย 
posted an update 12 days ago
view post
Post
2813
SPECK 2 IS ALREADY OUT: specklabs/Speck2-140M

Pretrained on 4x more tokens than the previous releases (20b vs 5b).
Instruct tuned versions are coming soon.
Very interesting models are coming soon too (hint: super long context).

Thanks for everyone supporting!
  • 3 replies
ยท
AtAndDevย 
posted an update 15 days ago
view post
Post
165
SPECK1.5 IS COMING SOON!
Same 5B token budget but much better corpus quality.

Also getting a ton of downloads, thanks for everyone downloading and liking <3

specklabs
AtAndDevย 
posted an update 18 days ago
view post
Post
145
NEW SPECK UPDATES:

Just hit #14 and #15 with out FIRST models on Open SLM Leaderboard. The models were trained on 5B tokens, while competing with similarly sized models trained on more than 6-20x the data.

A new base model Speck1.5-140M being trained right now on a higher quality corpus and will be released soon.
SpeckChat3 is coming very soon with 1 million samples, specifically designed to post train small base models.

Also, just to clarify stuff, we will NOT release anything that is NOT MIT licensed EVER. Openness is needed in small language research.

Thanks to everyone supporting the project, and stay tuned for new releases!
AtAndDevย 
posted an update 19 days ago
view post
Post
1879
SPECK UPDATES:
1 New instruct model tuned on top of Speck1-140M: specklabs/Speck1-140M-Instruct
2 Instruction tuning datasets
2 GGUFs

Much more coming soon:
Speck1.1-140M-Instruct that is post trained on SpeckChat2 will be coming very soon
New base model Speck1.5-140M is coming with a much higher quality corpus

Thanks to everyone who is already supporting the project, and stay tuned for new releases!
  • 3 replies
ยท
AtAndDevย 
posted an update 20 days ago
view post
Post
2118
FIRST SPECK MODEL RELEASED:
specklabs/Speck1-140M

new models coming very soon (both instruct and much better models), with much much higher training scale as i am getting marenostrum5 access soon!
we will be looking at 100b-2t token budgets :)
  • 4 replies
ยท
Q-bertย 
authored a paper about 1 month ago
suayptalhaย 
submitted a paper to Daily Papers about 1 month ago
suayptalhaย 
authored a paper about 1 month ago
suayptalhaย 
authored a paper 2 months ago
pankajpandey-devย 
posted an update 2 months ago
view post
Post
4146
๐Ÿ‡ฎ๐Ÿ‡ณ Qwen3.5-9B Hindi Instruct โ€” it stops thinking in English
Ask base Qwen3.5-9B a question in Hindi and it burns hundreds of tokens thinking in English inside its think block before a single Devanagari word appears โ€” then code-switches in the answer. I fine-tuned it to close the think block instantly and reply in pure, native Hindi.
โœ… Model (16-bit): pankajpandey-dev/qwen3.5-9b-hindi-instruct
โœ… GGUF (Q4/Q5/Q8): pankajpandey-dev/qwen3.5-9b-hindi-instruct-GGUF
โœ… Try it in the browser: pankajpandey-dev/qwen3.5-9b-hindi-demo
Recipe: Unsloth + LoRA (r=16, response-only loss) on 12.9k Hindi pairs โ€” AI4Bharat anudesh + dolly-hi + wikiHow-hi + Aya Hindi (human-written). The Q4_K_M is 5.4 GB and runs on a plain laptop CPU.
New in this run vs my earlier models: mixed in long-form native sources (wikiHow) after my last eval showed the fine-tune traded detail for conciseness โ€” this one keeps answers detailed and native.
Part of my weekly ๐Ÿ‡ฎ๐Ÿ‡ณ Hindi LLM Series. Feedback welcome ๐Ÿ™
#Hindi #IndicNLP #Qwen #GGUF #LocalLLM #Unsloth
  • 4 replies
ยท
pankajpandey-devย 
posted an update 3 months ago
view post
Post
7857
๐Ÿ‡ฎ๐Ÿ‡ณ New in my Hindi LLM Series: Gemma-4 E4B, fine-tuned for Hindi โ€” and it runs on your laptop's CPU.
I fine-tuned Google's new Gemma-4 E4B on ~10k Hindi instruction pairs (AI4Bharat: anudesh + dolly) using Unsloth + LoRA, on a single L4 GPU.
Then I ran an honest side-by-side eval: base Gemma-4 vs my fine-tune, across 25 Hindi prompts. The results were interesting ๐Ÿ‘‡
โœ… My fine-tune is more concise โ€” ask for "3 tips" and it gives exactly 3. Base writes a 1,200-character essay.

โœ… Pure native Hindi โ€” base keeps slipping into English ("เคธเค‚เคคเฅเคฒเคฟเคค เค†เคนเคพเคฐ (Eat a Balanced Diet)", "เคคเคพเคฐเคพ (Star)"). My fine-tune stays in clean Hindi.

โœ… Tighter instruction-following โ€” ask for a "short message" and it gives one, not a menu of options.
โš–๏ธ And to be honest: base Gemma-4 is more detailed and comprehensive. I didn't build a "smarter" model โ€” I built a focused, Hindi-native, edge-friendly one that runs as a 5GB GGUF (Q4) on CPU.
๐Ÿ”— Try it:

Live demo (CPU): pankajpandey-dev/gemma-4-e4b-hindi-demo
GGUF (Ollama/llama.cpp): pankajpandey-dev/gemma-4-e4b-hindi-instruct-GGUF
16-bit model: pankajpandey-dev/gemma-4-e4b-hindi-instruct

Built with @unsloth ยท Data by @ai4bharat ๐Ÿ™
#Hindi #LLM #Gemma #Unsloth #IndicNLP #GGUF
  • 12 replies
ยท
eienmojikiย 
posted an update 3 months ago
pankajpandey-devย 
posted an update 3 months ago
view post
Post
922
๐Ÿ‡ฎ๐Ÿ‡ณ Gemma-3-1B Hindi Instruct โ€” a Hindi LLM that runs fully offline, anywhere.
Last week I shipped Qwen3-4B Hindi. This week I went the other direction: how tiny can a useful Hindi model get? So I fine-tuned Gemma-3-1B on quality-filtered Hindi instruction data and shipped the full GGUF ladder.
โœ… Fine-tune (16-bit): pankajpandey-dev/gemma-3-1b-hindi-instruct
โœ… GGUF (Q4/Q5/Q8): pankajpandey-dev/gemma-3-1b-hindi-instruct-GGUF
Runs in Ollama, llama.cpp, and LM Studio. The Q4_K_M is just 806 MB โ€” runs on CPU, a cheap laptop, even a Raspberry Pi.
What I tried this round: chrF-filtered the training data to drop weak translations, and used response-only loss so the model learns how to answer, not how to repeat prompts.
Honest note: at 1B, Hindi fluency is strong but coherence is bounded by size โ€” it's a lightweight/edge experiment, not a 4B replacement. Gemma-3-4B Hindi is next.
Part of my Hindi LLM Series โ€” openly-licensed Indic models for local & edge use. Feedback welcome ๐Ÿ™
#Hindi #IndicNLP #GGUF #LocalLLM #Gemma #EdgeAI
pankajpandey-devย 
posted an update 4 months ago
view post
Post
15009
๐Ÿ‡ฎ๐Ÿ‡ณ Qwen3-4B Hindi Instruct v2 โ€” a Hindi LLM that runs on your own machine
Most strong Hindi-capable models are either huge or cloud-only. I wanted one that's small enough to run locally but actually follows instructions in Hindi โ€” so I fine-tuned Qwen3-4B on 10K Hindi instruction pairs and shipped it with a full GGUF quant ladder.
โœ… Fine-tune (16-bit): huggingface.co/pankajpandey-dev/Qwen3-4B-Hindi-Instruct-v2
โœ… GGUF (Q4/Q5/Q8): huggingface.co/pankajpandey-dev/Qwen3-4B-Hindi-Instruct-v2-GGUF
Runs in Ollama, llama.cpp, and LM Studio. The Q4_K_M is just 2.5 GB โ€” fits comfortably on a laptop, CPU or GPU.
Part of my Hindi LLM Series โ€” building openly-licensed Indic models for local and edge use. More coming (Gemma next). Feedback welcome ๐Ÿ™
#Hindi #IndicNLP #GGUF #LocalLLM #Qwen
  • 4 replies
ยท
pankajpandey-devย 
posted an update 4 months ago
view post
Post
702
๐Ÿ‡ฎ๐Ÿ‡ณ Just shipped: MiniCPM5-1B-Hindi-Instruct (+ GGUF quants)

First Hindi instruction-tuned fine-tune of OpenBMB's brand-new MiniCPM5-1B (released this week).

Trained with Unsloth + LoRA (r=32) on AI4Bharat's anudesh + dolly Hindi splits โ€” ~4k high-quality examples, 2 epochs on a single T4 in 60 minutes.

๐Ÿ”— Model (16-bit + LoRA adapter):
pankajpandey-dev/MiniCPM5-1B-Hindi-Instruct

๐Ÿ“ฆ GGUF quants for llama.cpp / Ollama / LM Studio:
pankajpandey-dev/MiniCPM5-1B-Hindi-Instruct-v1-GGUF

5 quant levels โ€” from Q3_K_M (~560 MB, runs on a Raspberry Pi) to Q8_0 (~1.2 GB, near-lossless). Q4_K_M is the recommended default.

Part of my ongoing ๐Ÿ‡ฎ๐Ÿ‡ณ Hindi LLM Series โ€” bringing strong open-source LLMs to Indian languages.

#Hindi #IndicNLP #MiniCPM5 #LoRA #Unsloth #GGUF #llamacpp #Ollama #LocalLLM
pankajpandey-devย 
posted an update 4 months ago
view post
Post
2706
๐Ÿงฌ Just uploaded K-quants of Carbon-3B for llama.cpp users!
@HuggingFaceBio released the original GGUF in bf16 only โ€” so I added the full quant ladder for CPU/edge inference:
โ€ข Q2_K โ†’ 1.4 GB
โ€ข Q3_K_M โ†’ 1.8 GB
โ€ข Q4_K_M โ†’ 2.1 GB โญ
โ€ข Q5_K_M โ†’ 2.4 GB
โ€ข Q6_K โ†’ 2.7 GB
โ€ข Q8_0 โ†’ 3.5 GB
๐Ÿ”— pankajpandey-dev/Carbon-3B-GGUF
Now you can generate DNA sequences on your laptop. Needs a llama.cpp build with PR #23410 (HybridDNATokenizer support).
Huge thanks to the HuggingFaceBio team for the original model ๐Ÿ™
#GGUF #llamacpp #genomics #DNA
pankajpandey-devย 
posted an update 4 months ago
view post
Post
281
Just released Qwen3-0.6B fine-tuned on Hindi instruction data ๐Ÿ‡ฎ๐Ÿ‡ณ

โœ… Full model: pankajpandey-dev/Qwen3-0.6B-Hindi-Instruct-v1
โœ… GGUF versions (Q2/Q4/Q5/Q8): pankajpandey-dev/Qwen3-0.6B-Hindi-Instruct-v1-GGUF

Smallest Hindi-capable GGUF โ€” runs on any laptop at 0.37GB.
Next: v2 with more data, better responses.

#Hindi #LLM #GGUF #OpenSource
Aurelien-Morganย 
posted an update 4 months ago
view post
Post
1146
@retrain-pipelines v0.2.0 is out !
I'm at Station F at My booth with GOSIM Paris 2026 today & tomorrow.
Come meet me for a live in-person demo and a chat !
  • 1 reply
ยท
suayptalhaย 
authored a paper 5 months ago