Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Thomas G. Lopes
commited on
Commit
·
f9fa0a5
1
Parent(s):
491aab8
routing mode for snippets
Browse files
src/lib/components/inference-playground/code-snippets.svelte
CHANGED
@@ -40,7 +40,7 @@
|
|
40 |
let showToken = $state(false);
|
41 |
|
42 |
type GetSnippetArgs = {
|
43 |
-
tokenStr
|
44 |
conversation: ConversationClass;
|
45 |
lang: InferenceSnippetLanguage;
|
46 |
};
|
@@ -53,6 +53,7 @@
|
|
53 |
max_tokens: data.config.max_tokens,
|
54 |
temperature: data.config.temperature,
|
55 |
top_p: data.config.top_p,
|
|
|
56 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
57 |
} as any;
|
58 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
@@ -61,7 +62,7 @@
|
|
61 |
}
|
62 |
|
63 |
if (isCustomModel(model)) {
|
64 |
-
const snippets = getInferenceSnippet(conversation, lang,
|
65 |
return snippets
|
66 |
.filter(s => s.client.startsWith("open") || lang === "sh")
|
67 |
.map(s => {
|
@@ -74,7 +75,7 @@
|
|
74 |
});
|
75 |
}
|
76 |
|
77 |
-
return getInferenceSnippet(conversation, lang,
|
78 |
}
|
79 |
|
80 |
// { javascript: 0, python: 0, http: 0 } at first
|
@@ -101,10 +102,10 @@
|
|
101 |
if (isCustomModel(conversation.model)) {
|
102 |
const t = conversation.model.accessToken;
|
103 |
|
104 |
-
return t && showToken ? t :
|
105 |
}
|
106 |
|
107 |
-
return token.value && showToken ? token.value :
|
108 |
});
|
109 |
|
110 |
const snippetsByLang = $derived({
|
|
|
40 |
let showToken = $state(false);
|
41 |
|
42 |
type GetSnippetArgs = {
|
43 |
+
tokenStr?: string;
|
44 |
conversation: ConversationClass;
|
45 |
lang: InferenceSnippetLanguage;
|
46 |
};
|
|
|
53 |
max_tokens: data.config.max_tokens,
|
54 |
temperature: data.config.temperature,
|
55 |
top_p: data.config.top_p,
|
56 |
+
accessToken: tokenStr,
|
57 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
58 |
} as any;
|
59 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
|
|
62 |
}
|
63 |
|
64 |
if (isCustomModel(model)) {
|
65 |
+
const snippets = getInferenceSnippet(conversation, lang, opts);
|
66 |
return snippets
|
67 |
.filter(s => s.client.startsWith("open") || lang === "sh")
|
68 |
.map(s => {
|
|
|
75 |
});
|
76 |
}
|
77 |
|
78 |
+
return getInferenceSnippet(conversation, lang, opts);
|
79 |
}
|
80 |
|
81 |
// { javascript: 0, python: 0, http: 0 } at first
|
|
|
102 |
if (isCustomModel(conversation.model)) {
|
103 |
const t = conversation.model.accessToken;
|
104 |
|
105 |
+
return t && showToken ? t : undefined;
|
106 |
}
|
107 |
|
108 |
+
return token.value && showToken ? token.value : undefined;
|
109 |
});
|
110 |
|
111 |
const snippetsByLang = $derived({
|
src/lib/utils/business.svelte.ts
CHANGED
@@ -316,8 +316,8 @@ export type GetInferenceSnippetReturn = InferenceSnippet[];
|
|
316 |
export function getInferenceSnippet(
|
317 |
conversation: ConversationClass,
|
318 |
language: InferenceSnippetLanguage,
|
319 |
-
accessToken: string,
|
320 |
opts?: {
|
|
|
321 |
messages?: ConversationEntityMembers["messages"];
|
322 |
streaming?: ConversationEntityMembers["streaming"];
|
323 |
max_tokens?: ConversationEntityMembers["config"]["max_tokens"];
|
@@ -342,7 +342,7 @@ export function getInferenceSnippet(
|
|
342 |
provider,
|
343 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
344 |
{ ...providerMapping, hfModelId: model.id } as any,
|
345 |
-
{ ...opts,
|
346 |
);
|
347 |
|
348 |
return allSnippets
|
|
|
316 |
export function getInferenceSnippet(
|
317 |
conversation: ConversationClass,
|
318 |
language: InferenceSnippetLanguage,
|
|
|
319 |
opts?: {
|
320 |
+
accessToken?: string;
|
321 |
messages?: ConversationEntityMembers["messages"];
|
322 |
streaming?: ConversationEntityMembers["streaming"];
|
323 |
max_tokens?: ConversationEntityMembers["config"]["max_tokens"];
|
|
|
342 |
provider,
|
343 |
// eslint-disable-next-line @typescript-eslint/no-explicit-any
|
344 |
{ ...providerMapping, hfModelId: model.id } as any,
|
345 |
+
{ ...opts, directRequest: false },
|
346 |
);
|
347 |
|
348 |
return allSnippets
|