Merge pull request #10373 from dannyl1u/logit_bias

feat: logit bias

backend/open_webui/utils/middleware.py

@@ -68,6 +68,7 @@ from open_webui.utils.misc import (
     get_last_user_message,
     get_last_assistant_message,
     prepend_to_first_user_message_content,
+    convert_logit_bias_input_to_json
 )
 from open_webui.utils.tools import get_tools
 from open_webui.utils.plugin import load_function_module_by_id
@@ -610,6 +611,11 @@ def apply_params_to_form_data(form_data, model):
 
         if "reasoning_effort" in params:
             form_data["reasoning_effort"] = params["reasoning_effort"]
+        if "logit_bias" in params:
+            try:
+                form_data["logit_bias"] = json.loads(convert_logit_bias_input_to_json(params["logit_bias"]))
+            except Exception as e:
+                print(f"Error parsing logit_bias: {e}")
 
     return form_data
 

backend/open_webui/utils/misc.py

@@ -6,6 +6,7 @@ import logging
 from datetime import timedelta
 from pathlib import Path
 from typing import Callable, Optional
+import json
 
 
 import collections.abc
@@ -450,3 +451,14 @@ def parse_ollama_modelfile(model_text):
         data["params"]["messages"] = messages
 
     return data
+
+def convert_logit_bias_input_to_json(user_input):
+    logit_bias_pairs = user_input.split(',')
+    logit_bias_json = {}
+    for pair in logit_bias_pairs:
+        token, bias = pair.split(':')
+        token = str(token.strip())
+        bias = int(bias.strip())
+        bias = 100 if bias > 100 else -100 if bias < -100 else bias
+        logit_bias_json[token] = bias
+    return json.dumps(logit_bias_json)

backend/open_webui/utils/payload.py

@@ -62,6 +62,7 @@ def apply_model_params_to_body_openai(params: dict, form_data: dict) -> dict:
         "reasoning_effort": str,
         "seed": lambda x: x,
         "stop": lambda x: [bytes(s, "utf-8").decode("unicode_escape") for s in x],
+        "logit_bias": lambda x: x,
     }
     return apply_model_params_to_body(params, form_data, mappings)
 

src/lib/components/chat/Settings/Advanced/AdvancedParams.svelte

@@ -17,6 +17,7 @@
 		stop: null,
 		temperature: null,
 		reasoning_effort: null,
+		logit_bias: null,
 		frequency_penalty: null,
 		repeat_last_n: null,
 		mirostat: null,
@@ -298,6 +299,49 @@
 		{/if}
 	</div>
 
+	<div class=" py-0.5 w-full justify-between">
+		<Tooltip
+			content={$i18n.t(
+				'Boosting or penalizing specific tokens for constrained responses. Bias values will be clamped between -100 and 100 (inclusive). (Default: none)'
+			)}
+			placement="top-start"
+			className="inline-tooltip"
+		>
+			<div class="flex w-full justify-between">
+				<div class=" self-center text-xs font-medium">
+					{$i18n.t('Logit Bias')}
+				</div>
+				<button
+					class="p-1 px-3 text-xs flex rounded-sm transition shrink-0 outline-hidden"
+					type="button"
+					on:click={() => {
+						params.logit_bias = (params?.logit_bias ?? null) === null ? '' : null;
+					}}
+				>
+					{#if (params?.logit_bias ?? null) === null}
+						<span class="ml-2 self-center"> {$i18n.t('Default')} </span>
+					{:else}
+						<span class="ml-2 self-center"> {$i18n.t('Custom')} </span>
+					{/if}
+				</button>
+			</div>
+		</Tooltip>
+
+		{#if (params?.logit_bias ?? null) !== null}
+			<div class="flex mt-0.5 space-x-2">
+				<div class=" flex-1">
+					<input
+						class="w-full rounded-lg pl-2 py-2 px-1 text-sm dark:text-gray-300 dark:bg-gray-850 outline-hidden"
+						type="text"
+						placeholder={$i18n.t('Enter comma-seperated "token:bias_value" pairs (example: 5432:100, 413:-100)')}
+						bind:value={params.logit_bias}
+						autocomplete="off"
+					/>
+				</div>
+			</div>
+		{/if}
+	</div>
+
 	<div class=" py-0.5 w-full justify-between">
 		<Tooltip
 			content={$i18n.t('Enable Mirostat sampling for controlling perplexity.')}

src/lib/components/chat/Settings/General.svelte

@@ -50,6 +50,7 @@
 		seed: null,
 		temperature: null,
 		reasoning_effort: null,
+		logit_bias: null,
 		frequency_penalty: null,
 		presence_penalty: null,
 		repeat_penalty: null,
@@ -348,6 +349,8 @@
 						temperature: params.temperature !== null ? params.temperature : undefined,
 						reasoning_effort:
 							params.reasoning_effort !== null ? params.reasoning_effort : undefined,
+						logit_bias:
+							params.logit_bias !== null ? params.logit_bias : undefined,
 						frequency_penalty:
 							params.frequency_penalty !== null ? params.frequency_penalty : undefined,
 						presence_penalty: