card: Parameters table first under H1 (sidebar badge empty by design — no weights)

README.md

@@ -30,6 +30,26 @@ tags:
 
 # Qwen3-Coder-Next-Spark-Agentic
 
+## Parameters
+
+| | |
+|--|--|
+| **Total** | **~80B** (HF count 79.67B BF16; non-embedding ~79B) |
+| **Active / token** | **~3B** |
+| **Context** | upstream card (see Qwen model page) |
+| **Arch** | MoE · `Qwen3NextForCausalLM` · 512 experts · 10 act · 1 shared |
+| **Default OS path** | Official **FP8** primary · GGUF alts secondary (weights not in this repo) |
+| **Source** | [Qwen/Qwen3-Coder-Next](https://huggingface.co/Qwen/Qwen3-Coder-Next) |
+| **This pack** | **runtime / probe pack only** — no weights here |
+
+> **Size is this table.** Hub sidebar **Parameters** badge stays empty on purpose — this repo has **no checkpoint weights** (pack / not-a-model).
+> Do not treat badge blank as “unknown size.” Numbers above are upstream / measured-pack facts.
+
+**Preview / Unverified** DGX Spark deployment pack for [Qwen/Qwen3-Coder-Next](https://huggingface.co/Qwen/Qwen3-Coder-Next) (~80B total / ~3B active).
+
+> This repository is a **runtime pack**, not a model checkpoint and not a re-quant.
+> **Harness format/routing probe** 2/2 (receipt only). **Not tool-reliability proof. Not ship / gate claim.** Verifier ≠ gate clearance.
+
 <!-- ainfera:honesty-block -->
 ## Honesty / launch class
 
@@ -45,29 +65,6 @@ tags:
 - **Default port:** `8001` (does not steal Laguna `:8000` unless pack *is* Laguna-S).
 - **Engine:** vLLM OpenAI server + official FP8 (`scripts/serve_vllm_fp8.sh`)
 
-
-
-## Parameters
-
-| | |
-|--|--|
-| **Total** | **~80B** (HF count 79.67B BF16; non-embedding ~79B) |
-| **Active / token** | **~3B** |
-| **Arch** | MoE · `Qwen3NextForCausalLM` · 512 experts · 10 act · 1 shared |
-| **Source** | [Qwen/Qwen3-Coder-Next](https://huggingface.co/Qwen/Qwen3-Coder-Next) |
-| **This pack** | **runtime / probe only** — no weights here |
-
-> HF native **Parameters** badge stays empty (no checkpoint in this repo) — size is the table above.
-> Native **Model tree** appears via YAML `base_model` for discoverability; pack is still runbook-only.
-> Full detail: [Model parameters](#model-parameters-upstream--not-this-pack).
-
-**Preview / Unverified** DGX Spark deployment pack for
-[Qwen/Qwen3-Coder-Next](https://huggingface.co/Qwen/Qwen3-Coder-Next)
-(~80B total / ~3B active).
-
-> This repository is a **runtime pack**, not a model checkpoint and not a re-quant.
-> **Harness format/routing probe** 2/2 (receipt only). **Not tool-reliability proof. Not ship / gate claim.** Verifier ≠ gate clearance.
-
 ## Status
 
 | | |
@@ -177,7 +174,6 @@ See `docs/WEIGHTS_LOCAL.md` + `SHA256SUMS`. Prefer **revision pins** over floati
 
 ---
 
-
 ---
 
 ## Probe quickstart (not a runnable day-0 ship path)