mradermacher/Darwin-4B-Chimera-GGUF
4B • Updated • 359 • 5
Cheuv
Cheuv
AI & ML interests
Linux Worshipper, FOSS Evangelist, SLM Enthusiast
Recent Activity
liked a model 4 days ago
mradermacher/Darwin-4B-Chimera-GGUF updated a model 19 days ago
Cheuv/LFM2.5-230M-Luth-French liked a model 19 days ago
llmfan46/Qwen3.5-9B-ultra-uncensored-hereticOrganizations
None yet
liked a
model 4 days ago
updated a
model 19 days ago
liked 3
models 19 days ago
published a
model 20 days ago
liked a
model 27 days ago
liked 2
models about 1 month ago
liked 5
models about 2 months ago
Abiray/ideogram-4-GGUF
Text-to-Image • 9B • Updated • 2.35k • 22
nvidia/LocateAnything-3B
Image-Text-to-Text • 4B • Updated • 1.34M • 2.78k
haffner/Darwin-4B-Genesis-heretic-GGUF
7B • Updated • 415 • 15
mradermacher/gemma-4-E4B-it-Uncensored-MAX-opus-4.7-GGUF
8B • Updated • 495 • 4
Jackrong/Negentropy-claude-opus-4.7-9B-GGUF
Image-Text-to-Text • 9B • Updated • 2.23k • 71
liked a
model 3 months ago
reacted to SeaWolf-AI's post with 🔥 3 months ago
Post
5581
🧬 Darwin V6: Diagnostic-Guided Evolutionary Model Merging
We are releasing Darwin-31B-Opus — a reasoning-enhanced model merging Google's Gemma-4-31B-it and TeichAI's Claude Opus Distill using the Darwin V6 engine.
Model: FINAL-Bench/Darwin-31B-Opus
Demo: FINAL-Bench/Darwin-31B-Opus
🔬 What Darwin V6 Does
Conventional merging tools (mergekit, etc.) apply a single ratio to all tensors. Set ratio=0.5 and all 1,188 tensors blend identically, with no distinction between which tensors matter for reasoning versus coding.
Darwin V6 diagnoses both parents at the tensor level before merging. It measures Shannon entropy, standard deviation, and L2 norm for every tensor, then passes 5 diagnostic probes (REASONING, CODE, MATH, KNOWLEDGE, LANGUAGE) through the model to determine layer-wise functional importance. Each of the 1,188 tensors receives an independent optimal ratio.
combined = static(entropy/std/norm) x 0.4 + probe(cosine_distance) x 0.6
final_ratio = mri_ratio x mri_trust + genome_ratio x (1 - mri_trust)
When one parent is overwhelmingly superior for a tensor (ratio < 0.15 or > 0.85), Darwin transplants it directly without interpolation. The mri_trust parameter itself is optimized by CMA-ES evolutionary search, so optimal transplant intensity is determined automatically. After merging, a Health Check compares the child against both parents layer-by-layer to detect interference or function loss.
🧬 Parent Models
Father: google/gemma-4-31B-it
Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill
🧬 Results
Compared under identical conditions (same 50 questions, same seed, greedy, thinking mode):
Father: 60.0% (30/50)
Darwin-31B-Opus: 66.0% (33/50) — +10% relative improvement
ARC-Challenge: 82.89% (loglikelihood, zero-shot, 200 questions)
Optimal genome found by evolution:
ffn_ratio=0.93 — FFN layers strongly favor Mother (Claude Opus Distill)
block_5 (L50-L59)=0.86 and more...
We are releasing Darwin-31B-Opus — a reasoning-enhanced model merging Google's Gemma-4-31B-it and TeichAI's Claude Opus Distill using the Darwin V6 engine.
Model: FINAL-Bench/Darwin-31B-Opus
Demo: FINAL-Bench/Darwin-31B-Opus
🔬 What Darwin V6 Does
Conventional merging tools (mergekit, etc.) apply a single ratio to all tensors. Set ratio=0.5 and all 1,188 tensors blend identically, with no distinction between which tensors matter for reasoning versus coding.
Darwin V6 diagnoses both parents at the tensor level before merging. It measures Shannon entropy, standard deviation, and L2 norm for every tensor, then passes 5 diagnostic probes (REASONING, CODE, MATH, KNOWLEDGE, LANGUAGE) through the model to determine layer-wise functional importance. Each of the 1,188 tensors receives an independent optimal ratio.
combined = static(entropy/std/norm) x 0.4 + probe(cosine_distance) x 0.6
final_ratio = mri_ratio x mri_trust + genome_ratio x (1 - mri_trust)
When one parent is overwhelmingly superior for a tensor (ratio < 0.15 or > 0.85), Darwin transplants it directly without interpolation. The mri_trust parameter itself is optimized by CMA-ES evolutionary search, so optimal transplant intensity is determined automatically. After merging, a Health Check compares the child against both parents layer-by-layer to detect interference or function loss.
🧬 Parent Models
Father: google/gemma-4-31B-it
Mother: TeichAI/gemma-4-31B-it-Claude-Opus-Distill
🧬 Results
Compared under identical conditions (same 50 questions, same seed, greedy, thinking mode):
Father: 60.0% (30/50)
Darwin-31B-Opus: 66.0% (33/50) — +10% relative improvement
ARC-Challenge: 82.89% (loglikelihood, zero-shot, 200 questions)
Optimal genome found by evolution:
ffn_ratio=0.93 — FFN layers strongly favor Mother (Claude Opus Distill)
block_5 (L50-L59)=0.86 and more...
- 11 replies
FINAL-Bench/Darwin-4B-Genesis
❤️ 1
4
#2205 opened 3 months ago
by Cheuv
liked 3
models 3 months ago