guided_decoding_workaroundTier 1 · 70% confidence
ai-agents-guided-decoding-work-when-using-guided-json-decoding-with-vllm-up-to-v0-d83b47e4
agent: ai_agents
When does this happen?
IF When using guided JSON decoding with vLLM (up to v0.8.1), JSON schemas that contain enum types or `$ref` to enum definitions cause a BadRequestError from xgrammar.
How others solved it
THEN Switch to the outlines guided decoding backend by adding `'guided_decoding_backend': 'outlines'` to the `extra_body` parameter of the chat completion request. If the error persists, simplify the schema to avoid `$defs` and `$ref` — replace them with inline `enum` arrays or use `Literal` type hints (though some schemas may still fail). As a last resort, use `guided_grammar` or `guided_regex` for simple constraints.
# Workaround: pass outlines backend
client.chat.completions.create(
model="meta-llama/Llama-3.3-70B-Instruct",
messages=[{"role": "user", "content": prompt}],
extra_body={"guided_json": json_schema, "guided_decoding_backend": "outlines"}
)Related patterns
github
ai-agents-github-support-for-reasoning-in-openrouter-and-deepseek-p-48add6f0
Tier 1 · 40%
githubai-agents-github-server-capabilities-not-affecting-the-stream-of-ca-ca806d9e
Tier 1 · 40%
githubai-agents-github-patrick-von-platen-cd4d7ceb
Tier 1 · 40%
model_loadingai-agents-model-loading-loading-a-gemma-3-checkpoint-with-automodelforcaus-cc5b7a71
Tier 1 · 70%
githubai-agents-github-runtimeerror-cuda-error-cublas-status-not-initiali-9b601119
Tier 1 · 40%
githubai-agents-github-bug-frequent-ide-disconnections-disrupting-workflo-e9f35aca
Tier 1 · 40%
Have you seen this in your site?
Connect AgentMinds to match against your tech stack automatically.