Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
2479.8
TFLOPS
30
92
Artem Darius Weber
milkomeda22
Follow
levvius's profile picture
DarkSteelDragon's profile picture
2 followers
·
28 following
AI & ML interests
None yet
Recent Activity
liked
a model
about 4 hours ago
microsoft/MediPhi
liked
a model
2 days ago
Jackrong/Gemopus-4-26B-A4B-it-GGUF
reacted
to
SeaWolf-AI
's
post
with 👀
5 days ago
Why This Matters — David Defeats Goliath MODEL: https://huggingface.co/FINAL-Bench/Darwin-4B-David SPACE: https://huggingface.co/spaces/FINAL-Bench/Darwin-4B-david We're releasing Darwin-4B-David, the first second-generation model in the Darwin Opus family. By evolving an already-evolved model, it achieves 85.0% on GPQA Diamond — surpassing its 58.6% original ancestor and even gemma-4-31B (84.3%) — with just 4.5B parameters. Second-Generation Evolution Most merges start from a base model and produce a single offspring. Darwin-4B-David breaks this pattern. The Father (Darwin-4B-Opus) was already evolved from gemma-4-E4B-it with Claude Opus reasoning distillation — a Gen-1 model. The Mother (DavidAU's DECKARD-Expresso-Universe) brings Unsloth deep tuning across 5 in-house datasets with thinking mode by default. Crossbreeding these two produced the first Gen-2 Darwin model. Darwin V6's Model MRI scanned both parents across all 42 layers, assigning independent optimal ratios per layer. The Mother's creativity and Korean language hotspot (Layer 22-25, weight 0.95) was maximally absorbed, while the Father's reasoning core (Layer 30-40, weight 0.48) was preserved. This is "Merge = Evolve" applied recursively — evolution of evolution. Benchmarks Darwin-4B-David scores 85.0% on GPQA Diamond (+26.4%p over original 58.6%), evaluated generatively with maj@8 (8 generations per question, majority vote), Epoch AI prompt format, thinking mode enabled, 50 sampled questions. On ARC-Challenge (25-shot, loglikelihood), both score 64.93% — expected, as loglikelihood doesn't capture thinking-mode reasoning differences. Why This Matters gemma-4-31B (30.7B) scores 84.3%. Darwin-4B-David surpasses it at 1/7th the size — no training, no RL, just 45 minutes of MRI-guided DARE-TIES on one H100. The name "David" honors Mother creator DavidAU and evokes David vs. Goliath.
View all activity
Organizations
milkomeda22
's activity
All
Models
Datasets
Spaces
Buckets
Papers
Collections
Community
Posts
Upvotes
Likes
Articles
liked
a model
about 4 hours ago
microsoft/MediPhi
Text Generation
•
4B
•
Updated
Dec 15, 2025
•
648
•
21
liked
a model
2 days ago
Jackrong/Gemopus-4-26B-A4B-it-GGUF
Text Generation
•
25B
•
Updated
7 days ago
•
36.1k
•
86
liked
a Space
5 days ago
Running
11
DeepResearch Bench
🔍
11
Explore and compare Deep Research models using benchmark rankings
liked
a model
7 days ago
zai-org/GLM-4.7-Flash
Text Generation
•
31B
•
Updated
Jan 29
•
717k
•
•
1.66k
liked
a model
8 days ago
Qwen/Qwen2.5-0.5B-Instruct
Text Generation
•
0.5B
•
Updated
Sep 25, 2024
•
5.6M
•
496
liked
a model
13 days ago
Skywork/Matrix-Game-3.0
Image-Text-to-Video
•
Updated
5 days ago
•
154
•
109
liked
a model
16 days ago
huizimao/gpt-oss-120b-uncensored-bf16
117B
•
Updated
Aug 11, 2025
•
93
•
10
liked
a model
22 days ago
facebook/tribev2
Updated
21 days ago
•
127k
•
400
liked
3 models
about 1 month ago
abilmansplus/whisper-turbo-kaz-rus-v1
Automatic Speech Recognition
•
Updated
Jan 14
•
1.61k
•
5
MiniMaxAI/MiniMax-M2.5
Text Generation
•
229B
•
Updated
Mar 10
•
916k
•
•
1.46k
Qwen/Qwen3.5-9B
Image-Text-to-Text
•
10B
•
Updated
Mar 2
•
6.22M
•
•
1.29k
liked
a model
about 2 months ago
allenai/olmOCR-2-7B-1025-FP8
Image-Text-to-Text
•
8B
•
Updated
Feb 19
•
278k
•
224
liked
a model
2 months ago
OpenGVLab/InternVL3_5-241B-A28B
Image-Text-to-Text
•
241B
•
Updated
Aug 29, 2025
•
533
•
138
liked
a model
3 months ago
Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice
Text-to-Speech
•
2B
•
Updated
Jan 29
•
1.49M
•
1.42k
liked
a Space
4 months ago
Running
3.21k
AnyCoder
🏆
3.21k
Generate code snippets with AI
liked
a model
4 months ago
mistralai/Devstral-Small-2-24B-Instruct-2512
24B
•
Updated
Feb 25
•
279k
•
586
liked
2 models
6 months ago
OpenGVLab/InternVideo2_5_Chat_8B
Video-Text-to-Text
•
8B
•
Updated
Aug 4, 2025
•
13k
•
89
ByteDance/Sa2VA-InternVL3-14B
Image-Text-to-Text
•
15B
•
Updated
Oct 16, 2025
•
15
•
9
liked
a model
7 months ago
Skywork/Matrix-Game-2.0
Image-to-Video
•
Updated
4 days ago
•
48
•
293
liked
a model
8 months ago
litagin/anime-whisper
Automatic Speech Recognition
•
0.8B
•
Updated
Nov 24, 2024
•
26.6k
•
133
Load more