oktayd/Qwen3.6-35B-v2-MoE-Ablit-Heretic-Uncensor-Hermes-MTP-Vision-Llama
Update README integrity hash
Add NOWPayments donation button
Clarify lineage sources for OBLITERATUS and Opus provenance
Simplify model-card navigation for grouped notes
Group supplemental model-card notes into nested dropdowns
Extend model lineage through v2 release
Add lineage sources and separate OBLITERATUS labels
Collapse meet-the-model sections into dropdowns
Add checkpoint summary and lineage integrity details
Expand compact model lineage name guide
Add top-level model card navigation
Inline training data summary in model card
Remove redundant edition chooser and prompt examples
Clarify runtime contexts in compact Ollama notes
Remove raw Ollama API request examples
Move laptop runtime status to benchmarks
Refresh README checksum
Remove redundant installation disclaimer
Collapse and unify benchmark diagnostics
Clarify release integrity manifests and practical limits
Remove redundant hardware link from introduction
Collapse quantization downloads into a compact panel
Present architecture and validation as a compact dropdown
Group setup reliability and hardware quickstarts into dropdowns
Group model architecture and personality under model introduction
Move architecture and quantization details below diagnostics
Compact diagnostics into scoped device dropdowns
Clarify one-step runtime prerequisites
Unify one-step runtime profile, personality, cache and loop guards
Add mandatory normal-runtime reliability profile and Windows installer
Complete planned 20-row IFBench runtime regression
Improve IFBench no-whitespace repair quality
Fail incomplete IFBench regression runs
Correct raw IFBench instruction rate
Add validated IFBench runtime regression and runner
Fix Q4 unattended runtime defaults and document validated repair path
Add nested Ubuntu and Windows installations inside all four hardware quickstarts
Add nested Ubuntu and Windows installations inside all four hardware quickstarts
Explain MoE and inherited stages; add per-test benchmark dropdowns and laptop capability roadmap
Explain MoE and inherited stages; add per-test benchmark dropdowns and laptop capability roadmap
Add collapsible H200, RTX 5090, laptop and experimental IQ2_M quickstarts
Add approved Istanbul banner and shorten dataset description
Refresh product-style card, personality, dataset credits and adult-content metadata
Publish Q36 v2 naming, complete edition guide and honest benchmark scope
Document completed Q4 llama.cpp diagnostic, including failures and unscored tasks
Finalize all seven validated quantization variants and import metadata
Upload SHA256SUMS with huggingface_hub
Upload GGUF-VALIDATION.json with huggingface_hub
Upload README.md with huggingface_hub
Add validated Q36-v1.3-IQ1_M.gguf