Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling
Paper
•
2508.16745
•
Published
•
27
AI & ML interests
None defined yet.
Recent Activity
b1l4lx1
authored
a
paper
13 days ago
romainhuet
updated
a
Space
27 days ago
romainhuet
published
a
Space
27 days ago
Post
4495
Excited to onboard FeatherlessAI on Hugging Face as an Inference Provider - they bring a fleet of 6,700+ LLMs on-demand on the Hugging Face Hub 🤯
Starting today, you'd be able to access all those LLMs (OpenAI compatible) on HF model pages and via OpenAI client libraries too! 💥
Go, play with it today: https://huggingface.co/blog/inference-providers-featherless
P.S. They're also bringing on more GPUs to support all your concurrent requests!
Starting today, you'd be able to access all those LLMs (OpenAI compatible) on HF model pages and via OpenAI client libraries too! 💥
Go, play with it today: https://huggingface.co/blog/inference-providers-featherless
P.S. They're also bringing on more GPUs to support all your concurrent requests!
- 1 reply
Post
4341
hey hey
@mradermacher
- VB from Hugging Face here, we'd love to onboard you over to our optimised xet backend! 💥
as you know we're in the process of upgrading our storage backend to xet (which helps us scale and offer blazingly fast upload/ download speeds too): https://huggingface.co/blog/xet-on-the-hub and now that we are certain that the backend can scale with even big models like Llama 4/ Qwen 3 - we;re moving to the next phase of inviting impactful orgs and users on the hub over as you are a big part of the open source ML community - we would love to onboard you next and create some excitement about it in the community too!
in terms of actual steps - it should be as simple as one of the org admins to join hf.co/join/xet - we'll take care of the rest.
p.s. you'd need to have a the latest hf_xet version of huggingface_hub lib but everything else should be the same: https://huggingface.co/docs/hub/storage-backends#using-xet-storage
p.p.s. this is fully backwards compatible so everything will work as it should! 🤗
as you know we're in the process of upgrading our storage backend to xet (which helps us scale and offer blazingly fast upload/ download speeds too): https://huggingface.co/blog/xet-on-the-hub and now that we are certain that the backend can scale with even big models like Llama 4/ Qwen 3 - we;re moving to the next phase of inviting impactful orgs and users on the hub over as you are a big part of the open source ML community - we would love to onboard you next and create some excitement about it in the community too!
in terms of actual steps - it should be as simple as one of the org admins to join hf.co/join/xet - we'll take care of the rest.
p.s. you'd need to have a the latest hf_xet version of huggingface_hub lib but everything else should be the same: https://huggingface.co/docs/hub/storage-backends#using-xet-storage
p.p.s. this is fully backwards compatible so everything will work as it should! 🤗
- 16 replies
mindhunter23
authored
a
paper
4 months ago
reach-vb
authored
a
paper
5 months ago
Post
2094
Hi community,
Few days back, I posted about my ongoing research on making reasoning mamba models and I found great insights from the community.
Today, I am announcing an update to the model weights. With newer checkpoints, the Falcon3 Mamba R1 model now outperforms very large transformer based LLMs (including Gemini) for Formal Logic questions of MMLU. It scores 60% on formal logic which is considered a tough subset of questions in MMLU.
I would highly appreciate your insights and suggestions on this new checkpoint.
Model Repo: hanzla/Falcon3-Mamba-R1-v0
Chat space: hanzla/Falcon3MambaReasoner
Few days back, I posted about my ongoing research on making reasoning mamba models and I found great insights from the community.
Today, I am announcing an update to the model weights. With newer checkpoints, the Falcon3 Mamba R1 model now outperforms very large transformer based LLMs (including Gemini) for Formal Logic questions of MMLU. It scores 60% on formal logic which is considered a tough subset of questions in MMLU.
I would highly appreciate your insights and suggestions on this new checkpoint.
Model Repo: hanzla/Falcon3-Mamba-R1-v0
Chat space: hanzla/Falcon3MambaReasoner
Post
4023
Hello community,
I want to share my work of creating a reasoning mamba model
I used GRPO over Falcon3 Mamba Instruct to make this model. It generates blazing fast response while building good logic to answer challenging questions.
Give it a try:
Model repo: hanzla/Falcon3-Mamba-R1-v0
Space: hanzla/Falcon3MambaReasoner
Looking forward to community feedback.
I want to share my work of creating a reasoning mamba model
I used GRPO over Falcon3 Mamba Instruct to make this model. It generates blazing fast response while building good logic to answer challenging questions.
Give it a try:
Model repo: hanzla/Falcon3-Mamba-R1-v0
Space: hanzla/Falcon3MambaReasoner
Looking forward to community feedback.
- 2 replies
Post
1267
Gemma 3 is a game changer for on device multimodal applications.
Try for yourself how a 4 billion parameter model can be so good.
hanzla/PlaygroundGemma3
Try for yourself how a 4 billion parameter model can be so good.
hanzla/PlaygroundGemma3
- 1 reply
mindhunter23
authored
2
papers
6 months ago
reach-vb
authored
a
paper
7 months ago
Post
7194
VLMs are going through quite an open revolution AND on-device friendly sizes:
1. Google DeepMind w/ PaliGemma2 - 3B, 10B & 28B: google/paligemma-2-release-67500e1e1dbfdd4dee27ba48
2. OpenGVLabs w/ InternVL 2.5 - 1B, 2B, 4B, 8B, 26B, 38B & 78B: https://huggingface.co/collections/OpenGVLab/internvl-25-673e1019b66e2218f68d7c1c
3. Qwen w/ Qwen 2 VL - 2B, 7B & 72B: Qwen/qwen2-vl-66cee7455501d7126940800d
4. Microsoft w/ FlorenceVL - 3B & 8B: @jiuhai
5. Moondream2 w/ 0.5B: https://huggingface.co/vikhyatk/
What a time to be alive! 🔥
1. Google DeepMind w/ PaliGemma2 - 3B, 10B & 28B: google/paligemma-2-release-67500e1e1dbfdd4dee27ba48
2. OpenGVLabs w/ InternVL 2.5 - 1B, 2B, 4B, 8B, 26B, 38B & 78B: https://huggingface.co/collections/OpenGVLab/internvl-25-673e1019b66e2218f68d7c1c
3. Qwen w/ Qwen 2 VL - 2B, 7B & 72B: Qwen/qwen2-vl-66cee7455501d7126940800d
4. Microsoft w/ FlorenceVL - 3B & 8B: @jiuhai
5. Moondream2 w/ 0.5B: https://huggingface.co/vikhyatk/
What a time to be alive! 🔥
Post
5032
Massive week for Open AI/ ML:
Mistral Pixtral & Instruct Large - ~123B, 128K context, multilingual, json + function calling & open weights
mistralai/Pixtral-Large-Instruct-2411
mistralai/Mistral-Large-Instruct-2411
Allen AI Tülu 70B & 8B - competive with claude 3.5 haiku, beats all major open models like llama 3.1 70B, qwen 2.5 and nemotron
allenai/tulu-3-models-673b8e0dc3512e30e7dc54f5
allenai/tulu-3-datasets-673b8df14442393f7213f372
Llava o1 - vlm capable of spontaneous, systematic reasoning, similar to GPT-o1, 11B model outperforms gemini-1.5-pro, gpt-4o-mini, and llama-3.2-90B-vision
Xkev/Llama-3.2V-11B-cot
Black Forest Labs Flux.1 tools - four new state of the art model checkpoints & 2 adapters for fill, depth, canny & redux, open weights
reach-vb/black-forest-labs-flux1-6743847bde9997dd26609817
Jina AI Jina CLIP v2 - general purpose multilingual and multimodal (text & image) embedding model, 900M params, 512 x 512 resolution, matroyoshka representations (1024 to 64)
jinaai/jina-clip-v2
Apple AIM v2 & CoreML MobileCLIP - large scale vision encoders outperform CLIP and SigLIP. CoreML optimised MobileCLIP models
apple/aimv2-6720fe1558d94c7805f7688c
apple/coreml-mobileclip
A lot more got released like, OpenScholar (https://huggingface.co/collections/OpenScholar/openscholar-v1-67376a89f6a80f448da411a6), smoltalk ( HuggingFaceTB/smoltalk), Hymba ( nvidia/hymba-673c35516c12c4b98b5e845f), Open ASR Leaderboard ( hf-audio/open_asr_leaderboard) and much more..
Can't wait for the next week! 🤗
Mistral Pixtral & Instruct Large - ~123B, 128K context, multilingual, json + function calling & open weights
mistralai/Pixtral-Large-Instruct-2411
mistralai/Mistral-Large-Instruct-2411
Allen AI Tülu 70B & 8B - competive with claude 3.5 haiku, beats all major open models like llama 3.1 70B, qwen 2.5 and nemotron
allenai/tulu-3-models-673b8e0dc3512e30e7dc54f5
allenai/tulu-3-datasets-673b8df14442393f7213f372
Llava o1 - vlm capable of spontaneous, systematic reasoning, similar to GPT-o1, 11B model outperforms gemini-1.5-pro, gpt-4o-mini, and llama-3.2-90B-vision
Xkev/Llama-3.2V-11B-cot
Black Forest Labs Flux.1 tools - four new state of the art model checkpoints & 2 adapters for fill, depth, canny & redux, open weights
reach-vb/black-forest-labs-flux1-6743847bde9997dd26609817
Jina AI Jina CLIP v2 - general purpose multilingual and multimodal (text & image) embedding model, 900M params, 512 x 512 resolution, matroyoshka representations (1024 to 64)
jinaai/jina-clip-v2
Apple AIM v2 & CoreML MobileCLIP - large scale vision encoders outperform CLIP and SigLIP. CoreML optimised MobileCLIP models
apple/aimv2-6720fe1558d94c7805f7688c
apple/coreml-mobileclip
A lot more got released like, OpenScholar (https://huggingface.co/collections/OpenScholar/openscholar-v1-67376a89f6a80f448da411a6), smoltalk ( HuggingFaceTB/smoltalk), Hymba ( nvidia/hymba-673c35516c12c4b98b5e845f), Open ASR Leaderboard ( hf-audio/open_asr_leaderboard) and much more..
Can't wait for the next week! 🤗
mindhunter23
authored
a
paper
10 months ago
Post
4764
What a brilliant week for Open Source AI!
Qwen 2.5 Coder by Alibaba - 0.5B / 1.5B / 3B / 7B / 14B/ 32B (Base + Instruct) Code generation LLMs, with 32B tackling giants like Gemnini 1.5 Pro, Claude Sonnet
Qwen/qwen25-coder-66eaa22e6f99801bf65b0c2f
LLM2CLIP from Microsoft - Leverage LLMs to train ultra-powerful CLIP models! Boosts performance over the previous SOTA by ~17%
microsoft/llm2clip-672323a266173cfa40b32d4c
Athene v2 Chat & Agent by NexusFlow - SoTA general LLM fine-tuned from Qwen 2.5 72B excels at Chat + Function Calling/ JSON/ Agents
Nexusflow/athene-v2-6735b85e505981a794fb02cc
Orca Agent Instruct by Microsoft - 1 million instruct pairs covering text editing, creative writing, coding, reading comprehension, etc - permissively licensed
microsoft/orca-agentinstruct-1M-v1
Ultravox by FixieAI - 70B/ 8B model approaching GPT4o level, pick any LLM, train an adapter with Whisper as Audio Encoder
reach-vb/ultravox-audio-language-model-release-67373b602af0a52b2a88ae71
JanusFlow 1.3 by DeepSeek - Next iteration of their Unified MultiModal LLM Janus with RectifiedFlow
deepseek-ai/JanusFlow-1.3B
Common Corpus by Pleais - 2,003,039,184,047 multilingual, commercially permissive and high quality tokens!
PleIAs/common_corpus
I'm sure I missed a lot, can't wait for the next week!
Put down in comments what I missed! 🤗
Qwen 2.5 Coder by Alibaba - 0.5B / 1.5B / 3B / 7B / 14B/ 32B (Base + Instruct) Code generation LLMs, with 32B tackling giants like Gemnini 1.5 Pro, Claude Sonnet
Qwen/qwen25-coder-66eaa22e6f99801bf65b0c2f
LLM2CLIP from Microsoft - Leverage LLMs to train ultra-powerful CLIP models! Boosts performance over the previous SOTA by ~17%
microsoft/llm2clip-672323a266173cfa40b32d4c
Athene v2 Chat & Agent by NexusFlow - SoTA general LLM fine-tuned from Qwen 2.5 72B excels at Chat + Function Calling/ JSON/ Agents
Nexusflow/athene-v2-6735b85e505981a794fb02cc
Orca Agent Instruct by Microsoft - 1 million instruct pairs covering text editing, creative writing, coding, reading comprehension, etc - permissively licensed
microsoft/orca-agentinstruct-1M-v1
Ultravox by FixieAI - 70B/ 8B model approaching GPT4o level, pick any LLM, train an adapter with Whisper as Audio Encoder
reach-vb/ultravox-audio-language-model-release-67373b602af0a52b2a88ae71
JanusFlow 1.3 by DeepSeek - Next iteration of their Unified MultiModal LLM Janus with RectifiedFlow
deepseek-ai/JanusFlow-1.3B
Common Corpus by Pleais - 2,003,039,184,047 multilingual, commercially permissive and high quality tokens!
PleIAs/common_corpus
I'm sure I missed a lot, can't wait for the next week!
Put down in comments what I missed! 🤗
Post
2039
Smol TTS models are here! OuteTTS-0.1-350M - Zero shot voice cloning, built on LLaMa architecture, CC-BY license! 🔥
> Pure language modeling approach to TTS
> Zero-shot voice cloning
> LLaMa architecture w/ Audio tokens (WavTokenizer)
> BONUS: Works on-device w/ llama.cpp ⚡
Three-step approach to TTS:
> Audio tokenization using WavTokenizer (75 tok per second)
> CTC forced alignment for word-to-audio token mapping
> Structured prompt creation w/ transcription, duration, audio tokens
The model is extremely impressive for 350M parameters! Kudos to the
OuteAI team on such a brilliant feat - I'd love to see this be applied on larger data and smarter backbones like SmolLM 🤗
Check out the models here: OuteAI/outetts-6728aa71a53a076e4ba4817c
> Pure language modeling approach to TTS
> Zero-shot voice cloning
> LLaMa architecture w/ Audio tokens (WavTokenizer)
> BONUS: Works on-device w/ llama.cpp ⚡
Three-step approach to TTS:
> Audio tokenization using WavTokenizer (75 tok per second)
> CTC forced alignment for word-to-audio token mapping
> Structured prompt creation w/ transcription, duration, audio tokens
The model is extremely impressive for 350M parameters! Kudos to the
OuteAI team on such a brilliant feat - I'd love to see this be applied on larger data and smarter backbones like SmolLM 🤗
Check out the models here: OuteAI/outetts-6728aa71a53a076e4ba4817c
reach-vb
authored
a
paper
10 months ago
Post
3208
Smol models ftw! AMD released AMD OLMo 1B - beats OpenELM, tiny llama on MT Bench, Alpaca Eval - Apache 2.0 licensed 🔥
> Trained with 1.3 trillion (dolma 1.7) tokens on 16 nodes, each with 4 MI250 GPUs
> Three checkpoints:
- AMD OLMo 1B: Pre-trained model
- AMD OLMo 1B SFT: Supervised fine-tuned on Tulu V2, OpenHermes-2.5, WebInstructSub, and Code-Feedback datasets
- AMD OLMo 1B SFT DPO: Aligned with human preferences using Direct Preference Optimization (DPO) on UltraFeedback dataset
Key Insights:
> Pre-trained with less than half the tokens of OLMo-1B
> Post-training steps include two-phase SFT and DPO alignment
> Data for SFT:
- Phase 1: Tulu V2
- Phase 2: OpenHermes-2.5, WebInstructSub, and Code-Feedback
> Model checkpoints on the Hub & Integrated with Transformers ⚡️
Congratulations & kudos to AMD on a brilliant smol model release! 🤗
amd/amd-olmo-6723e7d04a49116d8ec95070
> Trained with 1.3 trillion (dolma 1.7) tokens on 16 nodes, each with 4 MI250 GPUs
> Three checkpoints:
- AMD OLMo 1B: Pre-trained model
- AMD OLMo 1B SFT: Supervised fine-tuned on Tulu V2, OpenHermes-2.5, WebInstructSub, and Code-Feedback datasets
- AMD OLMo 1B SFT DPO: Aligned with human preferences using Direct Preference Optimization (DPO) on UltraFeedback dataset
Key Insights:
> Pre-trained with less than half the tokens of OLMo-1B
> Post-training steps include two-phase SFT and DPO alignment
> Data for SFT:
- Phase 1: Tulu V2
- Phase 2: OpenHermes-2.5, WebInstructSub, and Code-Feedback
> Model checkpoints on the Hub & Integrated with Transformers ⚡️
Congratulations & kudos to AMD on a brilliant smol model release! 🤗
amd/amd-olmo-6723e7d04a49116d8ec95070