exo-explore/exo · error · HTTPException

Failed to load model card: {exc}

Error message

Failed to load model card: {exc}

What it means

Error "Failed to load model card: {exc}" thrown in exo-explore/exo.

Source

Thrown at src/exo/api/main.py:524

        return placements[new_ids[0]]

    async def get_placement_previews(
        self,
        model_id: ModelId,
        node_ids: Annotated[list[NodeId] | None, Query()] = None,
    ) -> PlacementPreviewResponse:
        seen: set[tuple[ModelId, Sharding, InstanceMeta, int]] = set()
        previews: list[PlacementPreview] = []
        required_nodes = set(node_ids) if node_ids else None

        if len(list(self.state.topology.list_nodes())) == 0:
            return PlacementPreviewResponse(previews=[])

        try:
            model_card = await ModelCard.load(model_id)
        except Exception as exc:
            raise HTTPException(
                status_code=400, detail=f"Failed to load model card: {exc}"
            ) from exc
        instance_combinations: list[tuple[Sharding, InstanceMeta, int]] = []
        for sharding in (Sharding.Pipeline, Sharding.Tensor):
            for instance_meta in (InstanceMeta.MlxRing, InstanceMeta.MlxJaccl):
                instance_combinations.extend(
                    [
                        (sharding, instance_meta, i)
                        for i in range(
                            1, len(list(self.state.topology.list_nodes())) + 1
                        )
                    ]
                )
        # TODO: PDD
        # instance_combinations.append((Sharding.PrefillDecodeDisaggregation, InstanceMeta.MlxRing, 1))

        for sharding, instance_meta, min_nodes in instance_combinations:
            try:

View on GitHub (pinned to 21a54c5ea0)

When it happens

Trigger: Thrown at src/exo/api/main.py:524 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of exo-explore/exo@21a54c5ea0 (2026-08-26). Data as JSON: /api/errors/8df20c99e872df98. Report an issue: GitHub.