IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V2.1-GGUF
Show direct Ko-fi URL
Number navigation and refine optional support
Complete navigation index and remove duplicate support block
Move navigation to top and clarify optional support
Add optional Ko-fi support link
Correct fidelity tiers against measured delta perplexity
Add compact model-specific V2.1 comparison table
Add Qwen3.6 Abliterated V2.1 to lineup links
Replace generic Q5-Q6 redundancy with exact verified quality tier
Identify MiniPlus V2.1 edition in lineup links
Clarify established 35B MoE lineup and cross-links
Expand V2.1 model cross-promotion
Promote established V2.1 APEX-I-MiniPlus releases
Add system RAM inference guidance
Restore independent benchmark references; neutralize quantizer attribution
Remove third-party user references from model card
Keep WikiText-2 results while removing evaluation parameters
Remove exact RAM-offload and laptop throughput figures
Fix nested emphasis in Q5-Q6 fidelity statement
Emphasize Q5-Q6 fidelity at lower-than-Q3_K_M weight
Update model-specific Q3_K_M size reference (17.3 GB)
Tie Q5-Q6 fidelity callout to model result
Make quality comparison model-specific
Clarify external benchmark attribution
Separate author sampling recommendation from zephel01 llmbench setting & explicit Occamy V2 references
Replace model-specific Occamy failure example with generic checkpoint fidelity notice
Update model-specific Delta PPL and baseline perplexity values
Add third-party benchmark (zephel01), sampling parameters, and VRAM scaling
Add third-party benchmark (zephel01), sampling parameters, and VRAM scaling
Add third-party benchmark (zephel01), sampling parameters, and VRAM scaling
Add third-party benchmark (zephel01), sampling parameters, and VRAM scaling
Add third-party benchmark (zephel01), sampling parameters, and VRAM scaling
Add Quality Spectrum table comparing APEX-I-MiniPlus V2.1 against standard flat quantizations
Eliminate unintentional strikethrough by replacing bare tildes with explicit approx.
Fix markdown table formatting, resolve stray backticks, and correct PPL delta attribution
Fix perplexity wording redundancy and eliminate unrendered LaTeX math formatting
Add empirical WikiText-2 Perplexity (6.4725 +/- 0.16) and Sweet Spot analysis
Add standard mmproj-Q8_0.gguf for vision support
docs: clean excessive emojis across headings and callouts for professional styling
docs: clean formatting in important callout
docs: unify RAM streaming guide and eliminate repetitive text
docs: emphasize V2.1 system RAM execution with deep context, DDR4/DDR5 scaling, and VRAM offload
docs: clarify APEX-I-MiniPlus (all versions) vs generic community APEX-I-Mini
docs: add direct link to counterpart edition and Alder Lake CPU scaling testbed details
Streamline model card: restore audited Q8_0 attention gates, eliminate duplicate tables and duplicate benchmarks
Tailor Nex-N2.5-mini V2.1 model card for V1 lineage with transparent comparison
Update transparent Architecture Selection Guide: V1, V2, and V2.1 hardware comparison
Restore complete GPU Throughput Projections and 256K VRAM Scaling tables
Add Nex-N2.5-mini APEX-I-MiniPlus-V2.1 GGUF (13.74 GiB)
Initial README for Nex-N2.5-mini APEX-I-MiniPlus-V2.1