biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[V1] Structured Outputs + Thinking compatibility (#16577)

Browse Source 
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
This commit is contained in:
Aaron Pham
2025-05-14 18:45:24 -04:00
committed by GitHub
parent d93c976a0d
commit 2fc9075b82
10 changed files with 233 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/config.py
View File

@@ -2332,7 +2332,7 @@ class SpeculativeConfig:
`TypicalAcceptanceSampler`."""
speculative_token_tree: Optional[str] = None
"""Specifies the tree structure for speculative token generation.
"""Specifies the tree structure for speculative token generation.
"""
# required configuration params passed from engine
target_model_config: ModelConfig = field(default=None,
@@ -4024,7 +4024,7 @@ class VllmConfig:
"""LoRA configuration."""
speculative_config: Optional[SpeculativeConfig] = None
"""Speculative decoding configuration."""
decoding_config: Optional[DecodingConfig] = None
decoding_config: DecodingConfig = field(default_factory=DecodingConfig)
"""Decoding configuration."""
observability_config: Optional[ObservabilityConfig] = None
"""Observability configuration."""