Update README.md

README.md

@@ -19,10 +19,10 @@ LoRA merge analysis is hardware-agnostic — the same LoRA files always produce
 
 The AutoTuner computes pairwise conflict metrics (cosine similarity, sign conflicts, subspace overlap) and tests merge parameter combinations to find the best config for a set of LoRAs. These results are keyed by **content hash** (SHA256[:16] of file contents) — not by filename — so they're portable across systems and private by design.
 
-When `community_cache=download_only` or `upload_and_download` is set in the AutoTuner node:
+When `community_cache=upload_and_download` is set in the AutoTuner node:
 
 - **Download:** Before running analysis, the node checks this dataset for existing results. A config hit skips the entire sweep (~30–120s saved). Lora/pair cache hits speed up the analysis phase even without a full config hit.
-- **Upload:** After a successful sweep, results are uploaded if the local score beats the current community score for that LoRA set.
+- **Upload:** After a successful sweep (or when replaying from local memory), results are uploaded if the local score beats the current community score for that LoRA set.
 
 ---
 
@@ -54,31 +54,34 @@ All files include an `algo_version` field. Results from incompatible algorithm v
 
 ## Usage
 
-In the **LoRA AutoTuner** node, set `community_cache` to:
+In the **LoRA AutoTuner** node, set `community_cache` to `upload_and_download`. That's the only option — there's no passive download-only mode. If you benefit from the cache, you contribute back.
 
 | Value | Behavior |
 |-------|----------|
-| `disabled` | No network interaction (default) |
-| `download_only` | Download precomputed results anonymously — no account needed |
-| `upload_and_download` | Also contribute your results. Requires a Hugging Face account and `huggingface_hub` installed |
+| `disabled` (default) | No network interaction |
+| `upload_and_download` | Download precomputed results and contribute yours back |
 
 Network errors are silently ignored — the node always falls back to local computation.
 
 ---
 
-## Contributing
+## Setup
 
-**Setup (one time):**
+**One time:**
 
 ```bash
 pip install huggingface_hub
 huggingface-cli login
 ```
 
-That's it. The node picks up your stored token automatically.
+The node picks up your stored token automatically. No environment variables needed for most users.
 
-Alternatively, set `HF_TOKEN` as an environment variable (useful for headless/server setups).
+**Headless/server alternative:** set `HF_TOKEN` as an environment variable.
 
-**To contribute results:**
+**Then:** set `community_cache=upload_and_download` in the AutoTuner node and run as normal. Everything else is automatic.
 
-Set `community_cache=upload_and_download` in the AutoTuner node and run as normal. Results are uploaded automatically after each sweep if your score beats the current community score for that LoRA set. No other steps needed.
+---
+
+## Score-Based Replacement
+
+Configs are only uploaded when your local score beats the community score. Users with more thorough sweeps (`top_n=10`) or better hardware naturally contribute higher-quality results over time.