vllm-project/vllm · error · ValueError

Do not combine a positional recipe JSON source with --model/

Error message

Do not combine a positional recipe JSON source with --model/--hardware/--strategy discovery options.

What it means

The CLI accepts either a positional recipe JSON source OR discovery via --model/--hardware/--strategy, but not both. main() checks: if a positional source was given and any of the three discovery flags is also set, it raises this ValueError, because the two paths would disagree about which recipe to convert.

Source

Thrown at tools/recipes/recipe_json_to_vllm_config.py:634

    else:
        lines.append("# No recipe-specific environment variables.")

    lines.append("")
    Path(path).write_text("\n".join(lines), encoding="utf-8")
    Path(path).chmod(Path(path).stat().st_mode | 0o111)


def main() -> int:
    args = parse_args()

    try:
        source = args.source
        if source is None:
            source = discover_recipe_source(
                args.api_base, args.model, args.hardware, args.strategy
            )
        elif args.model or args.hardware or args.strategy:
            raise ValueError(
                "Do not combine a positional recipe JSON source with "
                "--model/--hardware/--strategy discovery options."
            )

        recipe = load_json(source)
        if not isinstance(recipe, dict):
            raise ValueError("Recipe JSON must be a JSON object.")

        argv = recipe_argv(recipe)
        config = argv_to_config(argv)
        write_config(args.config_out, source, recipe, config)
        write_env(args.env_out, source, recipe)
    except Exception as exc:
        print(f"ERROR: {exc}", file=sys.stderr)
        return 1

    print(f"Wrote {args.config_out}")
    print(f"Wrote {args.env_out}")

View on GitHub (pinned to c794754062)

Solutions

1. Drop either the positional source or all of --model/--hardware/--strategy from the command
2. If you have the JSON already, run: python tools/recipes/recipe_json_to_vllm_config.py recipe.json
3. If you want discovery, remove the positional path: python tools/recipes/recipe_json_to_vllm_config.py --model MODEL --hardware HW

Example fix

# before
python tools/recipes/recipe_json_to_vllm_config.py recipe.json --model Qwen/Qwen3-8B
# after
python tools/recipes/recipe_json_to_vllm_config.py recipe.json

if positional_source is not None and (args.model or args.hardware or args.strategy):
    raise SystemExit("pass either the positional source or --model/--hardware/--strategy, not both")

When it happens

Trigger: Running e.g. `recipe_json_to_vllm_config.py recipe.json --model Qwen/Qwen3-8B` — the positional file wins for content, but --model signals discovery, so the tool refuses the ambiguous invocation.

Common situations: Copy-pasting a previous discovery command and appending a downloaded JSON path; shell history editing; scripted invocations that always pass flags.

Related errors

• Unexpected level value ({args.level})
• Selected model {model.get('hf_id')!r} has no JSON API path.
• {model_json_url} did not return a JSON object.
• Model {model.get('hf_id')!r} has no rendered recommended_com
• Model {model.get('hf_id')!r} has no per-hardware renderings

AI-assisted analysis of vllm-project/vllm@c794754062 (2026-08-14). Data as JSON: /api/errors/e9744cde143ccd16. Report an issue: GitHub.