From 017d3704d2af9ff3bb61d20d506ed14278cdea06 Mon Sep 17 00:00:00 2001 From: Frank Date: Sat, 22 Aug 2026 15:06:51 -0700 Subject: [PATCH] Reorganize: llm/docs -> docs/LLM (LLM fleet knowledge base) The LLM fleet documentation (deployment runbook, model runbooks, benchmarks, research archive, retired-topology history) joins the project knowledge base as docs/LLM/. Wikilinks resolve by filename, so no link updates are needed for the move. --- {llm/docs => docs/LLM}/DFlash DSpark Speculative Decoding.md | 0 {llm/docs => docs/LLM}/Dankclaw Improvements.md | 0 {llm/docs => docs/LLM}/Dankclaw System Specs.md | 0 {llm/docs => docs/LLM}/Dankserver System Specs.md | 0 {llm/docs => docs/LLM}/FastFlowLM NPU Inference.md | 0 {llm/docs => docs/LLM}/Hybrid CPU GPU Processing Dankserver.md | 0 .../LLM}/Pipeline Parallelism/Ethernet Bandwidth Math.md | 0 .../LLM}/Pipeline Parallelism/Expected Performance.md | 0 .../LLM}/Pipeline Parallelism/Mixed-Vendor Experiments.md | 0 .../LLM}/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md | 0 .../LLM}/Pipeline Parallelism/Qwen3.8 PP Deployment.md | 0 {llm/docs => docs/LLM}/Pipeline Parallelism/Sample Script Set.md | 0 {llm/docs => docs/LLM}/Pipeline Parallelism/index.md | 0 {llm/docs => docs/LLM}/Pipeline Parallelism/llama.cpp RPC.md | 0 .../LLM}/Pipeline Parallelism/vLLM Pipeline Parallelism.md | 0 {llm/docs => docs/LLM}/Quantization by Hardware Architecture.md | 0 {llm/docs => docs/LLM}/RAM CPU Offloading in VLLM.md | 0 {llm/docs => docs/LLM}/Running Qwen3.6 27B FP8 on Dankclaw.md | 0 .../LLM}/Running Qwen3.6 35B A3B FP8 on Dankserver.md | 0 {llm/docs => docs/LLM}/Shared Memory Architecture Dankclaw.md | 0 {llm/docs => docs/LLM}/VLLM Index.md | 0 {llm/docs => docs/LLM}/VLLM Overview.md | 0 {llm/docs => docs/LLM}/ciru-activefpx-promptforge.md | 0 {llm/docs => docs/LLM}/dankserver-vram-growth.md | 0 {llm/docs => docs/LLM}/deployment.md | 0 {llm/docs => docs/LLM}/dflash2-large-vulkan.md | 0 {llm/docs => docs/LLM}/frank-default-large-benchmark.md | 0 {llm/docs => docs/LLM}/frank-default-small-benchmark.md | 0 {llm/docs => docs/LLM}/frank-redo-2026-08-17.md | 0 {llm/docs => docs/LLM}/index.md | 0 {llm/docs => docs/LLM}/kairic-edge.md | 0 {llm/docs => docs/LLM}/ling-3.0-flash-trial.md | 0 {llm/docs => docs/LLM}/ling-3.0-flash-vllm-plan.md | 0 {llm/docs => docs/LLM}/qwen3.8-spec-matrix.md | 0 {llm/docs => docs/LLM}/qwen3.8-strix-optimizations.md | 0 {llm/docs => docs/LLM}/vllm_research_findings.md | 0 36 files changed, 0 insertions(+), 0 deletions(-) rename {llm/docs => docs/LLM}/DFlash DSpark Speculative Decoding.md (100%) rename {llm/docs => docs/LLM}/Dankclaw Improvements.md (100%) rename {llm/docs => docs/LLM}/Dankclaw System Specs.md (100%) rename {llm/docs => docs/LLM}/Dankserver System Specs.md (100%) rename {llm/docs => docs/LLM}/FastFlowLM NPU Inference.md (100%) rename {llm/docs => docs/LLM}/Hybrid CPU GPU Processing Dankserver.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Ethernet Bandwidth Math.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Expected Performance.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Mixed-Vendor Experiments.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Qwen3.8 PP Deployment.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/Sample Script Set.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/index.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/llama.cpp RPC.md (100%) rename {llm/docs => docs/LLM}/Pipeline Parallelism/vLLM Pipeline Parallelism.md (100%) rename {llm/docs => docs/LLM}/Quantization by Hardware Architecture.md (100%) rename {llm/docs => docs/LLM}/RAM CPU Offloading in VLLM.md (100%) rename {llm/docs => docs/LLM}/Running Qwen3.6 27B FP8 on Dankclaw.md (100%) rename {llm/docs => docs/LLM}/Running Qwen3.6 35B A3B FP8 on Dankserver.md (100%) rename {llm/docs => docs/LLM}/Shared Memory Architecture Dankclaw.md (100%) rename {llm/docs => docs/LLM}/VLLM Index.md (100%) rename {llm/docs => docs/LLM}/VLLM Overview.md (100%) rename {llm/docs => docs/LLM}/ciru-activefpx-promptforge.md (100%) rename {llm/docs => docs/LLM}/dankserver-vram-growth.md (100%) rename {llm/docs => docs/LLM}/deployment.md (100%) rename {llm/docs => docs/LLM}/dflash2-large-vulkan.md (100%) rename {llm/docs => docs/LLM}/frank-default-large-benchmark.md (100%) rename {llm/docs => docs/LLM}/frank-default-small-benchmark.md (100%) rename {llm/docs => docs/LLM}/frank-redo-2026-08-17.md (100%) rename {llm/docs => docs/LLM}/index.md (100%) rename {llm/docs => docs/LLM}/kairic-edge.md (100%) rename {llm/docs => docs/LLM}/ling-3.0-flash-trial.md (100%) rename {llm/docs => docs/LLM}/ling-3.0-flash-vllm-plan.md (100%) rename {llm/docs => docs/LLM}/qwen3.8-spec-matrix.md (100%) rename {llm/docs => docs/LLM}/qwen3.8-strix-optimizations.md (100%) rename {llm/docs => docs/LLM}/vllm_research_findings.md (100%) diff --git a/llm/docs/DFlash DSpark Speculative Decoding.md b/docs/LLM/DFlash DSpark Speculative Decoding.md similarity index 100% rename from llm/docs/DFlash DSpark Speculative Decoding.md rename to docs/LLM/DFlash DSpark Speculative Decoding.md diff --git a/llm/docs/Dankclaw Improvements.md b/docs/LLM/Dankclaw Improvements.md similarity index 100% rename from llm/docs/Dankclaw Improvements.md rename to docs/LLM/Dankclaw Improvements.md diff --git a/llm/docs/Dankclaw System Specs.md b/docs/LLM/Dankclaw System Specs.md similarity index 100% rename from llm/docs/Dankclaw System Specs.md rename to docs/LLM/Dankclaw System Specs.md diff --git a/llm/docs/Dankserver System Specs.md b/docs/LLM/Dankserver System Specs.md similarity index 100% rename from llm/docs/Dankserver System Specs.md rename to docs/LLM/Dankserver System Specs.md diff --git a/llm/docs/FastFlowLM NPU Inference.md b/docs/LLM/FastFlowLM NPU Inference.md similarity index 100% rename from llm/docs/FastFlowLM NPU Inference.md rename to docs/LLM/FastFlowLM NPU Inference.md diff --git a/llm/docs/Hybrid CPU GPU Processing Dankserver.md b/docs/LLM/Hybrid CPU GPU Processing Dankserver.md similarity index 100% rename from llm/docs/Hybrid CPU GPU Processing Dankserver.md rename to docs/LLM/Hybrid CPU GPU Processing Dankserver.md diff --git a/llm/docs/Pipeline Parallelism/Ethernet Bandwidth Math.md b/docs/LLM/Pipeline Parallelism/Ethernet Bandwidth Math.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Ethernet Bandwidth Math.md rename to docs/LLM/Pipeline Parallelism/Ethernet Bandwidth Math.md diff --git a/llm/docs/Pipeline Parallelism/Expected Performance.md b/docs/LLM/Pipeline Parallelism/Expected Performance.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Expected Performance.md rename to docs/LLM/Pipeline Parallelism/Expected Performance.md diff --git a/llm/docs/Pipeline Parallelism/Mixed-Vendor Experiments.md b/docs/LLM/Pipeline Parallelism/Mixed-Vendor Experiments.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Mixed-Vendor Experiments.md rename to docs/LLM/Pipeline Parallelism/Mixed-Vendor Experiments.md diff --git a/llm/docs/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md b/docs/LLM/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md rename to docs/LLM/Pipeline Parallelism/Qwen3.6 27B Spec Methods.md diff --git a/llm/docs/Pipeline Parallelism/Qwen3.8 PP Deployment.md b/docs/LLM/Pipeline Parallelism/Qwen3.8 PP Deployment.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Qwen3.8 PP Deployment.md rename to docs/LLM/Pipeline Parallelism/Qwen3.8 PP Deployment.md diff --git a/llm/docs/Pipeline Parallelism/Sample Script Set.md b/docs/LLM/Pipeline Parallelism/Sample Script Set.md similarity index 100% rename from llm/docs/Pipeline Parallelism/Sample Script Set.md rename to docs/LLM/Pipeline Parallelism/Sample Script Set.md diff --git a/llm/docs/Pipeline Parallelism/index.md b/docs/LLM/Pipeline Parallelism/index.md similarity index 100% rename from llm/docs/Pipeline Parallelism/index.md rename to docs/LLM/Pipeline Parallelism/index.md diff --git a/llm/docs/Pipeline Parallelism/llama.cpp RPC.md b/docs/LLM/Pipeline Parallelism/llama.cpp RPC.md similarity index 100% rename from llm/docs/Pipeline Parallelism/llama.cpp RPC.md rename to docs/LLM/Pipeline Parallelism/llama.cpp RPC.md diff --git a/llm/docs/Pipeline Parallelism/vLLM Pipeline Parallelism.md b/docs/LLM/Pipeline Parallelism/vLLM Pipeline Parallelism.md similarity index 100% rename from llm/docs/Pipeline Parallelism/vLLM Pipeline Parallelism.md rename to docs/LLM/Pipeline Parallelism/vLLM Pipeline Parallelism.md diff --git a/llm/docs/Quantization by Hardware Architecture.md b/docs/LLM/Quantization by Hardware Architecture.md similarity index 100% rename from llm/docs/Quantization by Hardware Architecture.md rename to docs/LLM/Quantization by Hardware Architecture.md diff --git a/llm/docs/RAM CPU Offloading in VLLM.md b/docs/LLM/RAM CPU Offloading in VLLM.md similarity index 100% rename from llm/docs/RAM CPU Offloading in VLLM.md rename to docs/LLM/RAM CPU Offloading in VLLM.md diff --git a/llm/docs/Running Qwen3.6 27B FP8 on Dankclaw.md b/docs/LLM/Running Qwen3.6 27B FP8 on Dankclaw.md similarity index 100% rename from llm/docs/Running Qwen3.6 27B FP8 on Dankclaw.md rename to docs/LLM/Running Qwen3.6 27B FP8 on Dankclaw.md diff --git a/llm/docs/Running Qwen3.6 35B A3B FP8 on Dankserver.md b/docs/LLM/Running Qwen3.6 35B A3B FP8 on Dankserver.md similarity index 100% rename from llm/docs/Running Qwen3.6 35B A3B FP8 on Dankserver.md rename to docs/LLM/Running Qwen3.6 35B A3B FP8 on Dankserver.md diff --git a/llm/docs/Shared Memory Architecture Dankclaw.md b/docs/LLM/Shared Memory Architecture Dankclaw.md similarity index 100% rename from llm/docs/Shared Memory Architecture Dankclaw.md rename to docs/LLM/Shared Memory Architecture Dankclaw.md diff --git a/llm/docs/VLLM Index.md b/docs/LLM/VLLM Index.md similarity index 100% rename from llm/docs/VLLM Index.md rename to docs/LLM/VLLM Index.md diff --git a/llm/docs/VLLM Overview.md b/docs/LLM/VLLM Overview.md similarity index 100% rename from llm/docs/VLLM Overview.md rename to docs/LLM/VLLM Overview.md diff --git a/llm/docs/ciru-activefpx-promptforge.md b/docs/LLM/ciru-activefpx-promptforge.md similarity index 100% rename from llm/docs/ciru-activefpx-promptforge.md rename to docs/LLM/ciru-activefpx-promptforge.md diff --git a/llm/docs/dankserver-vram-growth.md b/docs/LLM/dankserver-vram-growth.md similarity index 100% rename from llm/docs/dankserver-vram-growth.md rename to docs/LLM/dankserver-vram-growth.md diff --git a/llm/docs/deployment.md b/docs/LLM/deployment.md similarity index 100% rename from llm/docs/deployment.md rename to docs/LLM/deployment.md diff --git a/llm/docs/dflash2-large-vulkan.md b/docs/LLM/dflash2-large-vulkan.md similarity index 100% rename from llm/docs/dflash2-large-vulkan.md rename to docs/LLM/dflash2-large-vulkan.md diff --git a/llm/docs/frank-default-large-benchmark.md b/docs/LLM/frank-default-large-benchmark.md similarity index 100% rename from llm/docs/frank-default-large-benchmark.md rename to docs/LLM/frank-default-large-benchmark.md diff --git a/llm/docs/frank-default-small-benchmark.md b/docs/LLM/frank-default-small-benchmark.md similarity index 100% rename from llm/docs/frank-default-small-benchmark.md rename to docs/LLM/frank-default-small-benchmark.md diff --git a/llm/docs/frank-redo-2026-08-17.md b/docs/LLM/frank-redo-2026-08-17.md similarity index 100% rename from llm/docs/frank-redo-2026-08-17.md rename to docs/LLM/frank-redo-2026-08-17.md diff --git a/llm/docs/index.md b/docs/LLM/index.md similarity index 100% rename from llm/docs/index.md rename to docs/LLM/index.md diff --git a/llm/docs/kairic-edge.md b/docs/LLM/kairic-edge.md similarity index 100% rename from llm/docs/kairic-edge.md rename to docs/LLM/kairic-edge.md diff --git a/llm/docs/ling-3.0-flash-trial.md b/docs/LLM/ling-3.0-flash-trial.md similarity index 100% rename from llm/docs/ling-3.0-flash-trial.md rename to docs/LLM/ling-3.0-flash-trial.md diff --git a/llm/docs/ling-3.0-flash-vllm-plan.md b/docs/LLM/ling-3.0-flash-vllm-plan.md similarity index 100% rename from llm/docs/ling-3.0-flash-vllm-plan.md rename to docs/LLM/ling-3.0-flash-vllm-plan.md diff --git a/llm/docs/qwen3.8-spec-matrix.md b/docs/LLM/qwen3.8-spec-matrix.md similarity index 100% rename from llm/docs/qwen3.8-spec-matrix.md rename to docs/LLM/qwen3.8-spec-matrix.md diff --git a/llm/docs/qwen3.8-strix-optimizations.md b/docs/LLM/qwen3.8-strix-optimizations.md similarity index 100% rename from llm/docs/qwen3.8-strix-optimizations.md rename to docs/LLM/qwen3.8-strix-optimizations.md diff --git a/llm/docs/vllm_research_findings.md b/docs/LLM/vllm_research_findings.md similarity index 100% rename from llm/docs/vllm_research_findings.md rename to docs/LLM/vllm_research_findings.md