rm max_completion_tokens

tlrmchlsmth · tlrmchlsmth · commit 7a0b1606bae4 · 2025-09-25T12:33:00.000-04:00
Signed-off-by: Tyler Michael Smith &lt;tyler@neuralmagic.com&gt;
diff --git a/src/guidellm/backend/openai.py b/src/guidellm/backend/openai.py
@@ -468,7 +468,6 @@ def _completions_payload(
                 max_output_tokens or self.max_output_tokens,
             )
             payload["max_tokens"] = max_output_tokens or self.max_output_tokens
-            payload["max_completion_tokens"] = payload["max_tokens"]
 
             if max_output_tokens:
                 # only set stop and ignore_eos if max_output_tokens set at request level

Original file line number	Diff line number	Diff line change
`@@ -468,7 +468,6 @@ def _completions_payload(`
`468`	`468`	`max_output_tokens or self.max_output_tokens,`
`469`	`469`	`)`
`470`	`470`	`payload["max_tokens"] = max_output_tokens or self.max_output_tokens`
`471`		`- payload["max_completion_tokens"] = payload["max_tokens"]`
`472`	`471`
`473`	`472`	`if max_output_tokens:`
`474`	`473`	`# only set stop and ignore_eos if max_output_tokens set at request level`