Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
1605.6
TFLOPS
76
7
338
ginipick
ginipick
Follow
AnkoUguisu's profile picture
menorki's profile picture
cxbz's profile picture
703 followers
ยท
156 following
AI & ML interests
None yet
Recent Activity
updated
a Space
20 days ago
ginipick/ai-news-daily
published
a Space
20 days ago
ginipick/ai-news-daily
reacted
to
SeaWolf-AI
's
post
with ๐
20 days ago
Why This Matters โ 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 โ surpassing its 58.6% original ancestor and even gemma-4-31B (84.3%) โ 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 โ 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 โ 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% โ 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 โ 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
ginipick
's datasets
9
Sort:ย Recently updated
ginipick/awesome-chatgpt-prompts
Viewer
โข
Updated
Nov 2, 2025
โข
203
โข
7
ginipick/Toucan-1.5M
Viewer
โข
Updated
Nov 2, 2025
โข
1.65M
โข
96
ginipick/finewiki
Viewer
โข
Updated
Nov 2, 2025
โข
61.4M
โข
21
ginipick/darwin-a2ap-analysis
Viewer
โข
Updated
Nov 2, 2025
โข
1k
โข
10
ginipick/market
Updated
Oct 25, 2025
โข
115
ginipick/darwin-a2ap-analysis-20250915_163155
Viewer
โข
Updated
Sep 15, 2025
โข
20
โข
189
ginipick/darwin-a2ap-analysis-20250915_151507
Viewer
โข
Updated
Sep 15, 2025
โข
100
โข
14
ginipick/pdf-test
Viewer
โข
Updated
May 28, 2025
โข
5
โข
9
ginipick/autotrain-data-autotrain-7u119-vc77x
Viewer
โข
Updated
May 9, 2024
โข
8
โข
16