crazy how we are speedruning LM fine-tuning before gta6 is even out. ๐คฏ
Daniel Fox PRO
FlameF0X
AI & ML interests
Pre-training text generator.
(Brother, im 18)
Please don't try to contact me.
Recent Activity
liked a model about 13 hours ago
surogate/Surogate-3.5-9B liked a model about 17 hours ago
baseten/GLM-5.2-Vision-NVFP4 upvoted a changelog 1 day ago
MCP Server EnhancementsOrganizations
reacted to Quazim0t0's post with ๐ 2 days ago
Post
1137
Two small model releases:
1. BE-ImgGen-Portrait-187M
BE = Byrne / Escarda - a text-to-image latent-diffusion DiT built entirely from the SpikeWhale / Escarda / Byrne family architecture, taken through six training campaigns to its current form: a painterly text-to-face generator with reliable coarse attribute conditioning (hair colour, beard, age, skin tone, freckles).
Final checkpoint: 294,236 steps ยท 162M DiT + 24M VAE ยท 55,030 5-point-aligned faces.
Repo: Quazim0t0/BE-ImgGen-Portrait-187M
2. CS2Dream - A CS2 Latent World Model (MIRA-style)
An action-conditioned world model for Counter-Strike 2 - feed it keyboard + mouse actions and it imagines the next frames. Built by adapting MIRA to stream RekaAI/CS2-10k (63 TB) with no local dataset copy.
Repo: Quazim0t0/CS2Dream
1. BE-ImgGen-Portrait-187M
BE = Byrne / Escarda - a text-to-image latent-diffusion DiT built entirely from the SpikeWhale / Escarda / Byrne family architecture, taken through six training campaigns to its current form: a painterly text-to-face generator with reliable coarse attribute conditioning (hair colour, beard, age, skin tone, freckles).
Final checkpoint: 294,236 steps ยท 162M DiT + 24M VAE ยท 55,030 5-point-aligned faces.
Repo: Quazim0t0/BE-ImgGen-Portrait-187M
2. CS2Dream - A CS2 Latent World Model (MIRA-style)
An action-conditioned world model for Counter-Strike 2 - feed it keyboard + mouse actions and it imagines the next frames. Built by adapting MIRA to stream RekaAI/CS2-10k (63 TB) with no local dataset copy.
Repo: Quazim0t0/CS2Dream
reacted to danielhanchen's post with ๐ 4 days ago
Post
5236
Gemma 4 is now faster and much more accurate! ๐
Google made huge improvements to tool-calling and chat accuracy, reliability + speed.
To get fixes, re-download our updated GGUF, MLX, NVFP4 quants!
Unsloth quants: https://huggingface.co/collections/unsloth/gemma-4
Gemma 4 Guide: https://unsloth.ai/docs/models/gemma-4
Google made huge improvements to tool-calling and chat accuracy, reliability + speed.
To get fixes, re-download our updated GGUF, MLX, NVFP4 quants!
Unsloth quants: https://huggingface.co/collections/unsloth/gemma-4
Gemma 4 Guide: https://unsloth.ai/docs/models/gemma-4
- 8 replies
reacted to cetusian's post with ๐ 18 days ago
Post
125
most labs won't bother teaching a small model to write and think in real romanian. no invented words, no english leaking in mid-sentence, diacritics intact.
we did.
surogate 3.5 is out. 2B and 4B, apache 2.0. invented word-forms cut from 4 per 1k down to ~1.5, and it reasons in the language you prompt it in, every time.
intelligence should speak your language too.
the models:
surogate/Surogate-3.5-4B
surogate/Surogate-3.5-2B
we did.
surogate 3.5 is out. 2B and 4B, apache 2.0. invented word-forms cut from 4 per 1k down to ~1.5, and it reasons in the language you prompt it in, every time.
intelligence should speak your language too.
the models:
surogate/Surogate-3.5-4B
surogate/Surogate-3.5-2B
posted an update 18 days ago
Post
196
Hello, people of Hugging Face!
I recently released FlameF0X/TinyMoE-100m-2x8-retrained, a small Mixture of Experts language model trained on the Smollm-Corpus. Built on top of the Mixtral architecture, itโs fully compatible with ๐ค Transformers right out of the box!
The model can produce somewhat coherent text on its own, and for some reason, it generates even more coherent responses when given a ChatLM template.
Iโm excited to see what you all come up with, and feel free to fine-tune it if youโd like. In the meantime, Iโll be working on developing the chat-trained version.
Demo: FlameF0X/TinyMoE-Playground
Collection: https://huggingface.co/collections/FlameF0X/tinymoe
I recently released FlameF0X/TinyMoE-100m-2x8-retrained, a small Mixture of Experts language model trained on the Smollm-Corpus. Built on top of the Mixtral architecture, itโs fully compatible with ๐ค Transformers right out of the box!
The model can produce somewhat coherent text on its own, and for some reason, it generates even more coherent responses when given a ChatLM template.
Iโm excited to see what you all come up with, and feel free to fine-tune it if youโd like. In the meantime, Iโll be working on developing the chat-trained version.
Demo: FlameF0X/TinyMoE-Playground
Collection: https://huggingface.co/collections/FlameF0X/tinymoe
reacted to ginigen-ai's post with โค๏ธ 23 days ago
Post
10464
๐ง Does your LLM know when it's about to be wrong?
Most leaderboards measure accuracy. We measure metacognition โ whether a model catches its own errors. Benchmark + leaderboard + adapters, all open. ๐
The surprise: even a K-AI #1 model (JGOS-31B-Citizen) is the strongest on multiple-choice traps (trap_rate 0.005 โ ~2 misses in 400) yet blind to its own free-form mistakes (self-confidence AUROC = 0.5, pure random). A tiny base-frozen adapter recovers that signal.
Two independent axes (never compared across a row): โ trap_rate โ does it fall for tempting trap options? (lower = stronger) โก adapter gain ฮ โ how much a lightweight adapter catches errors the model itself misses. (higher = more adapter value)
What's open: ๐ 300+100 trap problems (each with a hidden trap + TICOS type) ๐ 24-model leaderboard ๐งฉ 11 per-model adapters โ adapters, NOT fine-tunes (base stays frozen; the adapter just reads the hidden state โ P(wrong))
Submit any HF model โ auto-scored daily at 09:00 KST and added to the board.
๐ Leaderboard โ ginigen-ai/Metacognition-Leaderboard-Space
๐ Benchmark โ ginigen-ai/Metacognition-Bench
๐งฉ Adapters โ FINAL-Bench/metacognition-adapters-6a42c032e6beb803dd032961
๐ Article โ https://huggingface.co/blog/ginigen-ai/metacognition
Benchmark by ginigen-ai ยท Adapters by FINAL-Bench (Darwin/Chimera platform + AETHER metacognition tech).
Most leaderboards measure accuracy. We measure metacognition โ whether a model catches its own errors. Benchmark + leaderboard + adapters, all open. ๐
The surprise: even a K-AI #1 model (JGOS-31B-Citizen) is the strongest on multiple-choice traps (trap_rate 0.005 โ ~2 misses in 400) yet blind to its own free-form mistakes (self-confidence AUROC = 0.5, pure random). A tiny base-frozen adapter recovers that signal.
Two independent axes (never compared across a row): โ trap_rate โ does it fall for tempting trap options? (lower = stronger) โก adapter gain ฮ โ how much a lightweight adapter catches errors the model itself misses. (higher = more adapter value)
What's open: ๐ 300+100 trap problems (each with a hidden trap + TICOS type) ๐ 24-model leaderboard ๐งฉ 11 per-model adapters โ adapters, NOT fine-tunes (base stays frozen; the adapter just reads the hidden state โ P(wrong))
Submit any HF model โ auto-scored daily at 09:00 KST and added to the board.
๐ Leaderboard โ ginigen-ai/Metacognition-Leaderboard-Space
๐ Benchmark โ ginigen-ai/Metacognition-Bench
๐งฉ Adapters โ FINAL-Bench/metacognition-adapters-6a42c032e6beb803dd032961
๐ Article โ https://huggingface.co/blog/ginigen-ai/metacognition
Benchmark by ginigen-ai ยท Adapters by FINAL-Bench (Darwin/Chimera platform + AETHER metacognition tech).
- 11 replies
reacted to ginigen-ai's post with ๐ฅ 24 days ago
Post
5206
๐ณ The RoboCasa Kitchen Leaderboard
What does it take for a robot to handle kitchen chores the way a person does? It has to see (Vision), understand instructions (Language), and actually act (Action) โ and VLA (Vision-Language-Action) models are emerging as the answer. They're the bridge between large multimodal models and real-world embodied control.
RoboCasa Kitchen is a leading robot-learning benchmark in which a single-arm robot (Franka Panda) performs 24 atomic manipulation tasks โ picking up cups and bowls, opening drawers and doors, turning faucets, pressing buttons, and more โ inside a photorealistic simulated kitchen. Because the layout and object placement are randomized every episode, it tests genuine generalization rather than memorized motions. The score (success rate, SR) is the average fraction of the 24 tasks completed as instructed, measured over multiple seeds so results aren't down to luck.
The catch: this benchmark has no official leaderboard, and protocols (number of demonstrations, evaluation setup) differ from paper to paper, leaving scores scattered. Lining the numbers up naively quickly turns into an apples-to-oranges comparison.
This leaderboard fixes that by collecting published scores with their sources and comparing only what is genuinely comparable. It's split into three tables:
๐ Kitchen 24-task (matched) โ head-to-head under identical conditions (per the RLDX-1 Technical Report). This is the core ranking you can actually trust.
โ Other protocols โ self-reported under different setups (e.g. fewer demos). Not directly comparable, so kept separate.
๐ค GR1-Tabletop โ a different, humanoid-based variant suite, separated to avoid confusion.
Any researcher can submit their own model's score directly, and submissions are reviewed before they appear on the board. Every number links to its source paper, so you can verify it yourself.
๐ ginigen-ai/robocasa-kitchen-leaderboard
What does it take for a robot to handle kitchen chores the way a person does? It has to see (Vision), understand instructions (Language), and actually act (Action) โ and VLA (Vision-Language-Action) models are emerging as the answer. They're the bridge between large multimodal models and real-world embodied control.
RoboCasa Kitchen is a leading robot-learning benchmark in which a single-arm robot (Franka Panda) performs 24 atomic manipulation tasks โ picking up cups and bowls, opening drawers and doors, turning faucets, pressing buttons, and more โ inside a photorealistic simulated kitchen. Because the layout and object placement are randomized every episode, it tests genuine generalization rather than memorized motions. The score (success rate, SR) is the average fraction of the 24 tasks completed as instructed, measured over multiple seeds so results aren't down to luck.
The catch: this benchmark has no official leaderboard, and protocols (number of demonstrations, evaluation setup) differ from paper to paper, leaving scores scattered. Lining the numbers up naively quickly turns into an apples-to-oranges comparison.
This leaderboard fixes that by collecting published scores with their sources and comparing only what is genuinely comparable. It's split into three tables:
๐ Kitchen 24-task (matched) โ head-to-head under identical conditions (per the RLDX-1 Technical Report). This is the core ranking you can actually trust.
โ Other protocols โ self-reported under different setups (e.g. fewer demos). Not directly comparable, so kept separate.
๐ค GR1-Tabletop โ a different, humanoid-based variant suite, separated to avoid confusion.
Any researcher can submit their own model's score directly, and submissions are reviewed before they appear on the board. Every number links to its source paper, so you can verify it yourself.
๐ ginigen-ai/robocasa-kitchen-leaderboard
replied to AxionLab-official's post 25 days ago
i might a biased or not, but i think that the people behing "SupraLarps" might be racist. idk
surly is not the fact that they use the n word like 2 time
this is the profanity filter chrome extension btw.
reacted to AxionLab-official's post with โค๏ธ 25 days ago
Post
5078
โ ๏ธ Community Notice
We would like to clarify that SupraLabs has no affiliation, partnership, or connection whatsoever with "SupraLarps" or its members.
Please avoid interacting with their organization, repositories, or Spaces under the assumption that they are associated with us.
We are currently aware of the situation and have already contacted the appropriate channels to address it.
Thank you to everyone who continues to support SupraLabs. โค๏ธ
We would like to clarify that SupraLabs has no affiliation, partnership, or connection whatsoever with "SupraLarps" or its members.
Please avoid interacting with their organization, repositories, or Spaces under the assumption that they are associated with us.
We are currently aware of the situation and have already contacted the appropriate channels to address it.
Thank you to everyone who continues to support SupraLabs. โค๏ธ
- 12 replies
reacted to AxionLab-official's post with ๐ฅ about 1 month ago
Post
8529
Please, give a follow to SupraLabs!
We are researching the most, just to make the best medium models FOR YOU!
SupraLabs/Supra-A2A-Nano-Exp
SupraLabs/Supra-1.5-50M-Instruct-exp
SupraLabs/Supra-50M-Reasoning
SupraLabs/supra-title-50M-pre-gguf
Check more at Supralabs org!
SupraLabs
---
@LH-Tech-AI
@QyrouNnet-AI
@LyJonathan
@Mmorgan-ML
@User01110
We are researching the most, just to make the best medium models FOR YOU!
SupraLabs/Supra-A2A-Nano-Exp
SupraLabs/Supra-1.5-50M-Instruct-exp
SupraLabs/Supra-50M-Reasoning
SupraLabs/supra-title-50M-pre-gguf
Check more at Supralabs org!
---
@LH-Tech-AI
@QyrouNnet-AI
@LyJonathan
@Mmorgan-ML
@User01110
- 2 replies
reacted to Hari5115's post with ๐ฅ about 1 month ago
Post
1567
Bit addictive. Fair warning !!!
Chain combos, fever mode, daily leaderboard. Free, runs in your browser.
Beat the score if you can ๐ซง
๐ฎ Hari5115/neon-pop
#SendHelp #JustOneMoreGame #NeonPop #NotAddicted
Chain combos, fever mode, daily leaderboard. Free, runs in your browser.
Beat the score if you can ๐ซง
๐ฎ Hari5115/neon-pop
#SendHelp #JustOneMoreGame #NeonPop #NotAddicted
- 2 replies
New record ๐ฅ๐ฅ๐ฅ
reacted to Reubencf's post with ๐ฅ about 1 month ago
Post
3765
Shadows of Tomorrow is finally live on Hugging Face Spaces with Gradio.
Itโs a browser-playable RPG built with Godot, set in a post-nuclear future where players explore Magnus Province, collect medicinal plants, craft medicine, and help cure NPCs.
Play it here: Reubencf/Shadows_of_Tomorrow
Itโs a browser-playable RPG built with Godot, set in a post-nuclear future where players explore Magnus Province, collect medicinal plants, craft medicine, and help cure NPCs.
Play it here: Reubencf/Shadows_of_Tomorrow
- 11 replies
replied to their post about 1 month ago
Ah, MD doesn't work :(
posted an update about 1 month ago
Post
348
My models on the Intel Low-Bit LLM Leaderboard
Figured I'd share where my quantized models landed on Intel/low_bit_open_llm_leaderboard since I hadn't posted about it yet.
FlameF0X/Qwen3-4B-Distilled-Claude-4.6 (NVFP4 and MXFP4) sit at ranks 23 and 24 with 62.68% and 61.18% average, right below the base Qwen3-4B. Not bad considering they were distilled from Claude 4.6 rather than trained from scratch.
FlameF0X/LFM2.5-1.2B-Distilled-Claude-4.6 and FlameF0X/LFM2.5-1.2B-Thinking-CodeX land around rank 47-49, competitive with MiniCPM5-1B and the Qwen3 sub-1B models despite being a larger base architecture.
The funny one is FlameF0X/Qwen2-0.2B-pt and FlameF0X/Qwen2-0.2B-it. They're not properly trained โ genuinely undertrained, basically undefined โ and they still beat openai/gpt-oss-20b at rank 66. The 20B model. Not sure what that says but it's something.
FlameF0X/LFM2-Research is at the bottom of my lineup but it's a research artifact, not meant to be competitive.
Chart below showing my models vs nearby competitors, with size vs performance on the left.
Chart made by Claude
- 1 reply
reacted to their post with ๐ฅ about 2 months ago
reacted to wenhuach's post with ๐ฅ about 2 months ago
Post
4586
๐ We provide **free** hardware to quantize models at the [Intel Low Bit Open LLM Leaderboard]( Intel/low_bit_open_llm_leaderboard), currently supporting
โญ If you find it useful, please consider starring the AutoRound project on [GitHub](https://github.com/intel/auto-round)!
Pure RTN mode powered by AutoRoundโญ If you find it useful, please consider starring the AutoRound project on [GitHub](https://github.com/intel/auto-round)!
- 9 replies
reacted to pankajpandey-dev's post with ๐ฅ about 2 months ago
Post
2704
๐งฌ 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:
โข Q2_K โ 1.4 GB
โข Q3_K_M โ 1.8 GB
โข Q4_K_M โ 2.1 GB โญ
โข Q5_K_M โ 2.4 GB
โข Q6_K โ 2.7 GB
โข Q8_0 โ 3.5 GB
๐ 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 ๐
#GGUF #llamacpp #genomics #DNA
@HuggingFaceBio released the original GGUF in bf16 only โ so I added the full quant ladder for CPU/edge inference:
โข Q2_K โ 1.4 GB
โข Q3_K_M โ 1.8 GB
โข Q4_K_M โ 2.1 GB โญ
โข Q5_K_M โ 2.4 GB
โข Q6_K โ 2.7 GB
โข Q8_0 โ 3.5 GB
๐ 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 ๐
#GGUF #llamacpp #genomics #DNA
reacted to Crownelius's post with ๐ฅ 2 months ago
Post
4772
Howdy,
CompactAI-O is launching a tiny Model Golf, and the winner walks away with $50 in RunPod credits. Monthly. Every month. Show up, build, somebody wins.
What it is
Build the best language model you can under 100 million parameters, with at least a 1028-token context window. That's it. Any architecture, any tokenizer, any training scheme you can dream up at 3am. The only catch is it's gotta be open source (MIT, GPL, Apache, AGPL) take your pick.
It scratches the same itch as a Kaggle comp without the dataset\leaderboard nonsense. No fixed benchmark to game. No llama.cpp compatibility hoops. If you wanna train a 50M-param MoE with five experts and a tokenizer built on cookbooks, you can do that. Nothing stopping you.
The rules are listed in the discord and on the organization page if you're interested.
Why $50????
It's symbolic. It ain't gonna make anyone rich. But it's enough to cover a weekend of GPU time, enough to keep enthusiasts coming back, and not so much that it pulls in people who are just there for the money. Enthusiasts build interesting things. Interesting things move the field forward. A little incentive. I'd do it for $50 lol.
How to join
First round opens soon. Landing page is here:
โ https://huggingface.co/spaces/CompactAI-O/Tiny-model-golf
For questions or to swap ideas, the Discord's open:
โ https://discord.gg/y2jTct6Cxv
Excited to see what yall come up with. โฅ
โ Shane
CompactAI-O is launching a tiny Model Golf, and the winner walks away with $50 in RunPod credits. Monthly. Every month. Show up, build, somebody wins.
What it is
Build the best language model you can under 100 million parameters, with at least a 1028-token context window. That's it. Any architecture, any tokenizer, any training scheme you can dream up at 3am. The only catch is it's gotta be open source (MIT, GPL, Apache, AGPL) take your pick.
It scratches the same itch as a Kaggle comp without the dataset\leaderboard nonsense. No fixed benchmark to game. No llama.cpp compatibility hoops. If you wanna train a 50M-param MoE with five experts and a tokenizer built on cookbooks, you can do that. Nothing stopping you.
The rules are listed in the discord and on the organization page if you're interested.
Why $50????
It's symbolic. It ain't gonna make anyone rich. But it's enough to cover a weekend of GPU time, enough to keep enthusiasts coming back, and not so much that it pulls in people who are just there for the money. Enthusiasts build interesting things. Interesting things move the field forward. A little incentive. I'd do it for $50 lol.
How to join
First round opens soon. Landing page is here:
โ https://huggingface.co/spaces/CompactAI-O/Tiny-model-golf
For questions or to swap ideas, the Discord's open:
โ https://discord.gg/y2jTct6Cxv
Excited to see what yall come up with. โฅ
โ Shane
- 8 replies