Point at the public multimodalart/qwen3vl-conditioner, called without a token

app.py

@@ -2,7 +2,7 @@
 
 This Space holds the `transformer_ref` partition of the checkpoint and the two autoencoders, **unquantized
 bfloat16**, and nothing else. The 62.14 GiB Qwen3-VL conditioner lives in its own Space,
-[`minimax-h3-conditioner`](https://huggingface.co/spaces/diffusers-internal-dev/minimax-h3-conditioner), which this
+[`qwen3vl-conditioner`](https://huggingface.co/spaces/multimodalart/qwen3vl-conditioner), which this
 one calls over the gradio API for every request; what comes back is a safetensors file holding the two tensors the
 denoiser needs, `prompt_embeds` and `text_token_tags`.
 
@@ -28,7 +28,7 @@ import spaces
 import gradio as gr
 
 MODEL_REPO = os.environ.get("H3_MODEL_REPO", "diffusers-internal-dev/MiniMax-H3")
-CONDITIONER_SPACE = os.environ.get("H3_CONDITIONER", "diffusers-internal-dev/minimax-h3-conditioner")
+CONDITIONER_SPACE = os.environ.get("H3_CONDITIONER", "multimodalart/qwen3vl-conditioner")
 # `lazy` moves all 72.16 GiB onto the card on the first GPU call and leaves it there; `offload` hands placement to
 # `ComponentsManager.enable_auto_cpu_offload` instead. Neither puts anything on the card at *startup*, which is
 # deliberate — see `load_models`: the 150 GB storage quota, not the 95 GiB card, is what rules that out here.
@@ -305,7 +305,7 @@ def conditioner():
     if CLIENT is None:
         from gradio_client import Client
 
-        CLIENT = Client(CONDITIONER_SPACE, token=os.environ.get("HF_TOKEN"))
+        CLIENT = Client(CONDITIONER_SPACE)  # public Space, no org token: the request runs on the caller side quota
     return CLIENT