z-lab/Qwen3.6-35B-A3B-DFlash
Text Generation ā¢ 0.5B ā¢ Updated ā¢ 25.1k ā¢ 144
make
keisetsu
AI & ML interests
None yet
Recent Activity
liked a model about 2 hours ago
z-lab/Qwen3.6-35B-A3B-DFlash liked a model 7 days ago
deepsweet/Qwen3.6-35B-A3B-MLX-oQ8 liked a model 11 days ago
Jiunsong/supergemma4-26b-uncensored-gguf-v2Organizations
None yet
liked a
model about 2 hours ago
liked a
model 7 days ago
liked a
model 11 days ago
reacted to SeaWolf-AI's post with š„ 11 days ago
Post
5885
š§¬ Darwin-27B-Opus: 86.9% on GPQA Diamond ā World #5, Zero Training
We are excited to share Darwin-27B-Opus, a 27B model that achieved 86.9% on GPQA Diamond ā ranking #5 globally on the HuggingFace leaderboard ā without a single gradient update.
How? Darwin breeds pretrained models through evolutionary FFN crossbreeding. The father (Qwen3.5-27B) provides the reasoning architecture; the mother (Claude 4.6 Opus Reasoning Distilled) contributes structured chain-of-thought knowledge. CMA-ES automatically discovers optimal per-layer blending ratios ā no human tuning required.
The result surpasses the original Qwen3.5-27B (85.5%), GLM-5.1 (744B, 86.2%), and Qwen3.5-122B (86.6%). A 27B model outperforming 744B ā with zero training, zero data, one GPU, ~2 hours.
We also confirmed hybrid vigor on Korean benchmarks: Darwin-27B-KR (2nd generation offspring) surpassed both parents on CLIcK, winning 7 out of 11 categories. The evolutionary optimizer independently assigned 93% of FFN from the Korean-specialized mother while preserving 93% of attention from the reasoning-specialized father ā autonomously validating our core principle: FFN carries knowledge, Attention carries reasoning.
š Public release: 10 days ā 300+ community derivatives, 120K+ downloads.
š Links:
Darwin-27B-Opus: FINAL-Bench/Darwin-27B-Opus
article: https://huggingface.co/blog/FINAL-Bench/darwin-gpqa
Darwin Family Collection: https://huggingface.co/collections/FINAL-Bench/darwin-family
If foundation models are raw ore, Darwin is the forge. We are just getting started. š„
We are excited to share Darwin-27B-Opus, a 27B model that achieved 86.9% on GPQA Diamond ā ranking #5 globally on the HuggingFace leaderboard ā without a single gradient update.
How? Darwin breeds pretrained models through evolutionary FFN crossbreeding. The father (Qwen3.5-27B) provides the reasoning architecture; the mother (Claude 4.6 Opus Reasoning Distilled) contributes structured chain-of-thought knowledge. CMA-ES automatically discovers optimal per-layer blending ratios ā no human tuning required.
The result surpasses the original Qwen3.5-27B (85.5%), GLM-5.1 (744B, 86.2%), and Qwen3.5-122B (86.6%). A 27B model outperforming 744B ā with zero training, zero data, one GPU, ~2 hours.
We also confirmed hybrid vigor on Korean benchmarks: Darwin-27B-KR (2nd generation offspring) surpassed both parents on CLIcK, winning 7 out of 11 categories. The evolutionary optimizer independently assigned 93% of FFN from the Korean-specialized mother while preserving 93% of attention from the reasoning-specialized father ā autonomously validating our core principle: FFN carries knowledge, Attention carries reasoning.
š Public release: 10 days ā 300+ community derivatives, 120K+ downloads.
š Links:
Darwin-27B-Opus: FINAL-Bench/Darwin-27B-Opus
article: https://huggingface.co/blog/FINAL-Bench/darwin-gpqa
Darwin Family Collection: https://huggingface.co/collections/FINAL-Bench/darwin-family
If foundation models are raw ore, Darwin is the forge. We are just getting started. š„
liked a
model 16 days ago
liked a
model 6 months ago
liked 3
models 7 months ago
liked 5
models 9 months ago
openai/gpt-oss-120b
Text Generation ā¢ 120B ā¢ Updated ā¢ 3.65M ā¢ ā¢ 4.73k
openai/gpt-oss-20b
Text Generation ā¢ 22B ā¢ Updated ā¢ 6.59M ā¢ ā¢ 4.56k
MetaStoneTec/XBai-o4
33B ā¢ Updated ā¢ 15 ā¢ 192
Qwen/Qwen3-30B-A3B-Instruct-2507
Text Generation ā¢ Updated ā¢ 1.02M ā¢ ā¢ 801
SIP-med-LLM/SIP-jmed-llm-2-8x13b-OP-instruct
73B ā¢ Updated ā¢ 37 ā¢ 20
liked a
model 10 months ago
liked a
model 11 months ago
liked 3
models 12 months ago
liked a
model about 1 year ago
