docs: remove MIND-small ONNX from release card

README.md

@@ -22,7 +22,7 @@ lat/lon-only checkpoint is time-invariant. MINDSET contains the cached teacher e
 - `mind.pt` — fp32 PyTorch weights (453,967,275 bytes; pickle-based loading).
 - `mind.onnx` and `mind.onnx.data` — ONNX graph and external weights.
 - `mind.pt2` — PyTorch `ExportedProgram`.
-- `mind_small.safetensors`, `mind_small.onnx`, and `mind_small.pt2` — a 6.4M-parameter distilled student with a
+- `mind_small.safetensors` and `mind_small.pt2` — a 6.4M-parameter distilled student with a
   128-dimensional output.
 
 ## Usage
@@ -52,9 +52,8 @@ who have a compatible loader.
 
 ## MIND-small
 
-The repository contains `mind_small.safetensors`, `mind_small.onnx`, and `mind_small.pt2` for the 6.4M-parameter
-student. `mind_small.onnx.data` is not present in the live repository, so the ONNX student cannot be loaded from this
-release as an external-data graph. The safetensors and `.pt2` files require a compatible loader.
+The repository contains `mind_small.safetensors` and `mind_small.pt2` for the 6.4M-parameter
+student. These files require a compatible loader. An ONNX export of MIND-small is not currently available.
 
 ## Limitations