bartowski/Ling-3.0-flash-VL-GGUF
Image-Text-to-Text • 124B • Updated • 49 • 7
ddh0
ddh0
AI & ML interests
None yet
Recent Activity
liked a model about 21 hours ago
bartowski/Ling-3.0-flash-VL-GGUF liked a model 1 day ago
Danrisi/Canon_UltraReal liked a model 1 day ago
Danrisi/filmstills_qwen2.1Organizations
liked a
model about 21 hours ago
liked 7
models 1 day ago
Danrisi/Canon_UltraReal
Text-to-Image • Updated • 713 • 10
Danrisi/filmstills_qwen2.1
Text-to-Image • Updated • 98 • 4
Danrisi/samsung_qwen2.1
Text-to-Image • Updated • 84 • 2
Danrisi/nicegirls2_qwen2.1
Text-to-Image • Updated • 113 • 2
Danrisi/lenovo_qwen2.1
Text-to-Image • Updated • 103 • 4
Danrisi/canon_qwen2.1
Text-to-Image • Updated • 88 • 1
Danrisi/grainscape_qwen2.1
Text-to-Image • Updated • 74 • 1
upvoted an article 1 day ago
Article
Transformers now runs llama.cpp quants
- +1
marcsun13, ArthurZ, lysandre
• • 62 liked 2
models 3 days ago
liked a
model 4 days ago
upvoted a collection 4 days ago
liked a
model 4 days ago
liked a
model 5 days ago
liked a
Space 7 days ago
liked a
model 7 days ago
reacted to eaddario's post with ❤️🔥 10 days ago
Post
115
Experimental global target bits‑per‑weight quantization of XHToken/Spark-X2.5-1.7B and XHToken/Spark-X2.5-4B.
Unlike standard llama.cpp quantization that rely on fixed type heuristics (e.g., Q4_K_M), the Target BPW approach automatically optimizes per-tensor precision where it matters the most, and produces high quality models that meet a precise global size target.
Key Advantages:
- VRAM Maximization: Can generate high quality models sized exactly to fit hardware constraints (e.g., fitting the model into exactly 24GB VRAM).
- Data-Driven Precision: Quantization mix is determined by actual weight error sensitivity rather than hardcoded rules, often yielding better PPL/KLD size trade-offs.
Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's card.
eaddario/Spark-X2.5-1.7B-GGUF
eaddario/Spark-X2.5-4B-GGUF
Unlike standard llama.cpp quantization that rely on fixed type heuristics (e.g., Q4_K_M), the Target BPW approach automatically optimizes per-tensor precision where it matters the most, and produces high quality models that meet a precise global size target.
Key Advantages:
- VRAM Maximization: Can generate high quality models sized exactly to fit hardware constraints (e.g., fitting the model into exactly 24GB VRAM).
- Data-Driven Precision: Quantization mix is determined by actual weight error sensitivity rather than hardcoded rules, often yielding better PPL/KLD size trade-offs.
Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's card.
eaddario/Spark-X2.5-1.7B-GGUF
eaddario/Spark-X2.5-4B-GGUF
updated a
model 10 days ago