frank-default-small: revert to Huihui FP8 — FreeToken v0.1.2 cannot load compressed-tensors FP8

Two live-verified loader failures on Dankserver (2026-09-01), each a
clean exit at load (no swap-thrash, 37 GB fits RAM):

1. ornith-ai/Ornith-1.5-35B-A3B-FP8 (compressed-tensors PER-CHANNEL fp8):
   ValueError: Missing MoE expert source layers: {'gate_up': [0, 1, ...]}
   — nonstandard expert tensor naming, provider finds no expert sources.

2. lamianlbe/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-fp8
   (compressed-tensors BLOCK-fp8 128x128, standard per-expert naming):
   RuntimeError: The size of tensor a (2048) must match the size of
   tensor b (...) in _dequant_fp8_weight — the dequant path multiplies
   weight * weight_scale directly and cannot expand compressed-tensors
   block scales.

Empirical matrix for FreeToken v0.1.2 on qwen3_5_moe (RTX 3090):
  modelopt block-fp8 (quant_method 'fp8' + weight_scale_inv)  -> LOADS
  compressed-tensors per-channel FP8                          -> fails
  compressed-tensors block-fp8 128x128                        -> fails

batsclamp/Huihui-...-abliterated-FP8 is the proven uncensored checkpoint;
restoring it until a modelopt/NVFP4 requant of a wanted model exists or
FreeToken learns compressed-tensors.

scripts/frank-default-small.env

@@ -1,45 +1,30 @@
-# Tunables for frank-default-small (lamianlbe/Qwen3.6-35B-A3B-Uncensored-
-# HauhauCS-Aggressive-fp8, qwen3_5_moe hybrid MoE, Dankserver, RTX 3090).
+# Tunables for frank-default-small (batsclamp/Huihui-Qwen3.6-35B-A3B-
+# abliterated-FP8, qwen3_5_moe hybrid MoE, Dankserver, RTX 3090).
 #
 # Runtime: FreeToken v0.1.2 (edge-native MoE serving engine).
 # Installed via uv into /home/dank/freetoken/.venv (CPython 3.12).
 # FreeToken keeps expert weights in host RAM with an LRU GPU cache;
-# KV is paged with radix prefix sharing.  Only the full-attention
-# layers page KV (~20 KiB/token); the GDN layers use constant-size
+# KV is paged with radix prefix sharing.  Only the 10 full-attention
+# layers page KV (~20 KiB/token); the 30 GDN layers use constant-size
 # per-sequence state (mamba slot pool).
 #
-# Model: lamianlbe/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-fp8
-# (HuggingFace safetensors, ~37.5 GB on disk).  HauhauCS "Aggressive"
-# uncensored finetune of Qwen3.6-35B-A3B — same qwen3_5_moe arch as the
-# Huihui FP8 it replaces (40 layers: 10 full-attention + 30 GDN, 256
-# experts / 8 used, 2 KV heads × 128 dim, native 262144 context) and the
-# same ~35 GB FP8 weight mass, so the FreeToken memory/pool plan carries
-# unchanged.  Quantized to compressed-tensors BLOCK-fp8 (e4m3, dynamic
-# group-128 activations, 128×128 weight blocks) with the standard
-# per-expert tensor layout (mlp.experts.{N}.gate_proj/up_proj/down_proj
-# + .weight_scale) that FreeToken's MoE loader reads — the layout the
-# Ornith compressed-tensors checkpoint lacked.  Also ships a BF16 visual
-# encoder + MTP module (both in the quant "ignore" list; text-only
-# serving ignores them).  The RTX 3090 has no FP8 tensor cores, so FP8
-# runs via upcast.  Replaced batsclamp/Huihui-...-FP8 on 2026-09-01.
-#
-# NOTE (loader risk): compressed-tensors block-fp8 is not the modelopt
-# block-fp8 the runbook lists as ✅ — it shares the block layout and the
-# per-expert naming but carries a different quant_method.  Load is
-# verified live; 37.5 GB fits RAM so a failure is a clean exit, never a
-# swap-thrash.
-#
-# Chat template: the checkpoint ships its own chat_template.jinja (the
-# format this finetune was trained on); FreeToken reads it from the model
-# directory — no override.  The fleet Qwen-Sharp template is deliberately
-# NOT forced here: applying a template the model was not trained on risks
-# format drift on an uncensored finetune.  To force Sharp, set
-# CHAT_TEMPLATE_FILE=/home/dank/serverAppConfigs/scripts/chat-template.jinja.
+# Model: batsclamp/Huihui-Qwen3.6-35B-A3B-abliterated-FP8 (HuggingFace
+# safetensors).  Abliterated (refusal-removed) Huihui finetune of
+# Qwen3.6-35B-A3B, quantized to the SAME native format the previous
+# official checkpoint used: modelopt block-fp8 e4m3, dynamic
+# activations, 128x128 weight_block_size, bf16 weight_scale_inv;
+# lm_head/embeds/norms/router gates/visual/MTP kept BF16.  Un-fused
+# per-expert layout matches the official Qwen FP8, so the FreeToken
+# loader path is identical; ~35 GB on disk.  Ships the fixed Qwen3.6
+# chat template (froggeric/Qwen-Fixed-Chat-Templates) — FreeToken
+# reads it from the checkpoint; no template override needed.
+# Replaced Qwen/Qwen3.6-35B-A3B-FP8 on 2026-08-30; pool numbers
+# unchanged — same arch, same weight mass.
 #
 # Pool sizing: GPU budget (~17.6 GiB after weights + fixed caches) is
 # split MoE-first.  MOE_CACHE_SIZE pins resident expert slots; NUM_TOKENS
 # pins the KV pool.  At 2048 slots / 453005 tokens the split is ~20%
-# residency and ~8.6 GiB KV.  Live resize via
+# residency (2048/10240) and ~8.6 GiB KV.  Live resize via
 # `ft ctl cache --kv N --moe M` can grow KV by shrinking MoE without
 # restart (the reverse requires a restart).
 #
@@ -47,8 +32,8 @@
 # The ROCm HIP runtime reads those bare env vars (see the 2026-08-13 TEMP
 # incident).  This box is NVIDIA/CUDA, but the rule keeps the env portable.
 
-# Model directory (safetensors checkpoint, ~37.5 GB, compressed-tensors block-fp8).
-MODEL_DIR=/home/dank/models/hauhaucs-qwen36-35b-a3b-aggressive-fp8
+# Model directory (safetensors checkpoint, ~35 GB, block-fp8).
+MODEL_DIR=/home/dank/models/huihui-qwen36-35b-a3b-abliterated-fp8
 
 # FreeToken venv (uv-managed CPython 3.12, ft CLI, flashinfer, torch+cu130).
 FT_VENV=/home/dank/freetoken/.venv

scripts/frank-default-small.service

@@ -1,5 +1,5 @@
 [Unit]
-Description=FreeToken Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-FP8 qwen3_5_moe hybrid MoE (dankserver, frank-default-small)
+Description=FreeToken Huihui-Qwen3.6-35B-A3B-abliterated-FP8 qwen3_5_moe hybrid MoE (dankserver, frank-default-small)
 After=network-online.target
 Wants=network-online.target
 

scripts/serve-frank-small.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
-# Serve lamianlbe/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-fp8 as
-# frank-default-small via FreeToken on Dankserver (RTX 3090, 62 GB RAM).
+# Serve batsclamp/Huihui-Qwen3.6-35B-A3B-abliterated-FP8 as
+# frank-default-small via FreeToken on Dankserver (RTX 3090, 60 GB RAM).
 #
 # Runtime: FreeToken v0.1.2 (edge-native MoE serving engine).
 #   - Experts live in host RAM; an LRU cache of expert slots is resident
@@ -10,15 +10,14 @@
 #     hold constant-size per-sequence state (mamba slots).
 #   - OpenAI-compatible API (/v1/chat/completions, /v1/models).
 #
-# Model: lamianlbe/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-fp8 —
-# HauhauCS "Aggressive" uncensored finetune of Qwen3.6-35B-A3B in
-# compressed-tensors block-fp8 (e4m3, dynamic group-128 activations,
-# 128×128 weight blocks) with the standard per-expert tensor layout
-# (mlp.experts.{N}.gate_proj/up_proj/down_proj + .weight_scale).  Same
-# qwen3_5_moe arch and ~35 GB FP8 mass as the Huihui FP8 it replaces: 40
-# layers (10 full-attention + 30 GDN), 256 experts / 8 used, 2 KV heads ×
-# 128 dim, native 262144 context.  FreeToken reads the checkpoint's own
-# chat_template.jinja from the model directory (no template override).
+# Model: batsclamp/Huihui-Qwen3.6-35B-A3B-abliterated-FP8 — Huihui
+# abliterated finetune of Qwen3.6-35B-A3B in the same native modelopt
+# block-fp8 format (e4m3, 128x128 weight_block_size, un-fused per-expert
+# layout) as the official FP8 checkpoint this service ran before; the
+# FreeToken loader path is unchanged.  40 layers (10 attention + 30
+# GDN), 256 experts / 8 used, 2 KV heads × 256 dim.  ~35 GB on disk.
+# The checkpoint ships the fixed Qwen3.6 chat template; FreeToken reads
+# it from the model directory.
 #
 # Pool sizing: the GPU budget is split MoE-first between expert residency
 # and KV pages.  --moe-cache-size pins the number of resident expert
@@ -40,7 +39,7 @@ if [[ -f "$ENV_FILE" ]]; then
 fi
 
 # --- defaults (used when the env file does not set them) -----------------
-: "${MODEL_DIR:=/home/dank/models/hauhaucs-qwen36-35b-a3b-aggressive-fp8}"
+: "${MODEL_DIR:=/home/dank/models/huihui-qwen36-35b-a3b-abliterated-fp8}"
 : "${ALIAS:=frank-default-small}"
 : "${BIND_ADDR:=0.0.0.0}"
 : "${LISTEN_PORT:=7770}"