sgl-project/sglang · error · ValueError
MiniMax H3 AdaLN cache must be built on CUDA
Error message
MiniMax H3 AdaLN cache must be built on CUDA
What it means
Aborts the MiniMax H3 AdaLN cache build when no CUDA device is available, because the kernels used to build the cache are CUDA-only.
Solutions
- Run on a machine with a working GPU and CUDA toolchain
- Pass --device cuda (or a specific cuda:N)
- If GPU is present, fix the environment (driver, CUDA-visible-devices, container runtime) so torch.cuda.is_available() returns True
Example fix
# before --device cpu # after --device cuda
Defensive patterns
Strategy: validation
Validate before calling
import torch
if not torch.cuda.is_available():
raise SystemExit("CUDA required; no GPU visible") Prevention
- Gate GPU-only tooling behind an availability check in wrappers
- Verify nvidia-smi and container GPU passthrough before launching
When it happens
Trigger: Running the tool with --device cpu, or on a machine/container where no usable CUDA device is visible.
Common situations: Building the cache on a CPU-only box or inside a container without GPU passthrough; CUDA driver mismatch making torch.cuda.is_available() False; typo in the device string.
Related errors
- cuda.bindings.driver is required for CUDA VMM operations
- no supported CUDA VMM allocation handle type
- A CUDA VMM-enabled model must provide a multimodal processor
- AdaLN cache must cover at least one timestep plan
- All tensors must have the same data type
AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28).
Data as JSON: /api/errors/0e9a840aca00c1db.
Report an issue: GitHub.
Appendix: source
Thrown at python/sglang/multimodal_gen/tools/build_minimax_h3_adaln_cache.py:159
*,
weight_map: dict[str, str],
files: dict[str, Any],
device: torch.device,
) -> torch.Tensor:
tensor_file = files[weight_map[name]]
return tensor_file.get_tensor(name).to(device)
def main() -> None:
args = _parse_args()
if args.num_inference_steps < 2 and args.timesteps is None:
raise ValueError("--num-inference-steps must be at least 2")
mode_variant = _MODE_VARIANTS[args.mode]
if args.model_variant != mode_variant:
raise ValueError(f"--mode {args.mode} requires {mode_variant}")
device = torch.device(args.device)
if device.type != "cuda" or not torch.cuda.is_available():
raise ValueError("MiniMax H3 AdaLN cache must be built on CUDA")
index_path = args.transformer_path / "model.safetensors.index.json"
with index_path.open() as f:
weight_map = json.load(f)["weight_map"]
plans = _cache_timestep_plans(args)
if not plans or any(plan.numel() == 0 for plan in plans):
raise ValueError("AdaLN cache must cover at least one timestep plan")
max_plan_length = max(plan.numel() for plan in plans)
plan_timesteps = torch.zeros((len(plans), max_plan_length), dtype=torch.float32)
plan_lengths = torch.tensor([plan.numel() for plan in plans], dtype=torch.int64)
block_params = torch.empty(
(len(plans), max_plan_length, _NUM_BLOCKS, _BLOCK_PARAM_WIDTH),
dtype=torch.bfloat16,
)
final_params = torch.empty(
(len(plans), max_plan_length, _FINAL_PARAM_WIDTH), dtype=torch.bfloat16
)View on GitHub (pinned to 0132848349)