@huggingface/inference 3.9.0 → 3.9.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (9) hide show
  1. package/dist/index.cjs +3 -27
  2. package/dist/index.js +3 -27
  3. package/dist/src/lib/getInferenceProviderMapping.d.ts +2 -2
  4. package/dist/src/lib/getInferenceProviderMapping.d.ts.map +1 -1
  5. package/dist/src/providers/fal-ai.d.ts.map +1 -1
  6. package/package.json +3 -3
  7. package/src/lib/getInferenceProviderMapping.ts +1 -17
  8. package/src/providers/fal-ai.ts +0 -10
  9. package/src/snippets/templates.exported.ts +1 -1
package/dist/index.cjs CHANGED
@@ -342,15 +342,7 @@ var FalAITextToImageTask = class extends FalAITask {
342
342
  ...omit(params.args, ["inputs", "parameters"]),
343
343
  ...params.args.parameters,
344
344
  sync_mode: true,
345
- prompt: params.args.inputs,
346
- ...params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath ? {
347
- loras: [
348
- {
349
- path: buildLoraPath(params.mapping.hfModelId, params.mapping.adapterWeightsPath),
350
- scale: 1
351
- }
352
- ]
353
- } : void 0
345
+ prompt: params.args.inputs
354
346
  };
355
347
  if (params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath) {
356
348
  payload.loras = [
@@ -1323,7 +1315,7 @@ function getProviderHelper(provider, task) {
1323
1315
 
1324
1316
  // package.json
1325
1317
  var name = "@huggingface/inference";
1326
- var version = "3.9.0";
1318
+ var version = "3.9.2";
1327
1319
 
1328
1320
  // src/providers/consts.ts
1329
1321
  var HARDCODED_MODEL_INFERENCE_MAPPING = {
@@ -1388,22 +1380,6 @@ async function getInferenceProviderMapping(params, options) {
1388
1380
  `Model ${params.modelId} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
1389
1381
  );
1390
1382
  }
1391
- if (providerMapping.adapter === "lora") {
1392
- const treeResp = await (options?.fetch ?? fetch)(`${HF_HUB_URL}/api/models/${params.modelId}/tree/main`);
1393
- if (!treeResp.ok) {
1394
- throw new Error(`Unable to fetch the model tree for ${params.modelId}.`);
1395
- }
1396
- const tree = await treeResp.json();
1397
- const adapterWeightsPath = tree.find(({ type, path }) => type === "file" && path.endsWith(".safetensors"))?.path;
1398
- if (!adapterWeightsPath) {
1399
- throw new Error(`No .safetensors file found in the model tree for ${params.modelId}.`);
1400
- }
1401
- return {
1402
- ...providerMapping,
1403
- hfModelId: params.modelId,
1404
- adapterWeightsPath
1405
- };
1406
- }
1407
1383
  return { ...providerMapping, hfModelId: params.modelId };
1408
1384
  }
1409
1385
  return null;
@@ -2303,7 +2279,7 @@ var templates = {
2303
2279
  "basicAudio": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n data,\n model: "{{ model.id }}",\n provider: "{{ provider }}",\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(output);',
2304
2280
  "basicImage": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n data,\n model: "{{ model.id }}",\n provider: "{{ provider }}",\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(output);',
2305
2281
  "conversational": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst chatCompletion = await client.chatCompletion({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(chatCompletion.choices[0].message);',
2306
- "conversationalStream": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nlet out = "";\n\nconst stream = await client.chatCompletionStream({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nfor await (const chunk of stream) {\n if (chunk.choices && chunk.choices.length > 0) {\n const newContent = chunk.choices[0].delta.content;\n out += newContent;\n console.log(newContent);\n } \n}',
2282
+ "conversationalStream": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nlet out = "";\n\nconst stream = client.chatCompletionStream({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nfor await (const chunk of stream) {\n if (chunk.choices && chunk.choices.length > 0) {\n const newContent = chunk.choices[0].delta.content;\n out += newContent;\n console.log(newContent);\n }\n}',
2307
2283
  "textToImage": `import { InferenceClient } from "@huggingface/inference";
2308
2284
 
2309
2285
  const client = new InferenceClient("{{ accessToken }}");
package/dist/index.js CHANGED
@@ -285,15 +285,7 @@ var FalAITextToImageTask = class extends FalAITask {
285
285
  ...omit(params.args, ["inputs", "parameters"]),
286
286
  ...params.args.parameters,
287
287
  sync_mode: true,
288
- prompt: params.args.inputs,
289
- ...params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath ? {
290
- loras: [
291
- {
292
- path: buildLoraPath(params.mapping.hfModelId, params.mapping.adapterWeightsPath),
293
- scale: 1
294
- }
295
- ]
296
- } : void 0
288
+ prompt: params.args.inputs
297
289
  };
298
290
  if (params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath) {
299
291
  payload.loras = [
@@ -1266,7 +1258,7 @@ function getProviderHelper(provider, task) {
1266
1258
 
1267
1259
  // package.json
1268
1260
  var name = "@huggingface/inference";
1269
- var version = "3.9.0";
1261
+ var version = "3.9.2";
1270
1262
 
1271
1263
  // src/providers/consts.ts
1272
1264
  var HARDCODED_MODEL_INFERENCE_MAPPING = {
@@ -1331,22 +1323,6 @@ async function getInferenceProviderMapping(params, options) {
1331
1323
  `Model ${params.modelId} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
1332
1324
  );
1333
1325
  }
1334
- if (providerMapping.adapter === "lora") {
1335
- const treeResp = await (options?.fetch ?? fetch)(`${HF_HUB_URL}/api/models/${params.modelId}/tree/main`);
1336
- if (!treeResp.ok) {
1337
- throw new Error(`Unable to fetch the model tree for ${params.modelId}.`);
1338
- }
1339
- const tree = await treeResp.json();
1340
- const adapterWeightsPath = tree.find(({ type, path }) => type === "file" && path.endsWith(".safetensors"))?.path;
1341
- if (!adapterWeightsPath) {
1342
- throw new Error(`No .safetensors file found in the model tree for ${params.modelId}.`);
1343
- }
1344
- return {
1345
- ...providerMapping,
1346
- hfModelId: params.modelId,
1347
- adapterWeightsPath
1348
- };
1349
- }
1350
1326
  return { ...providerMapping, hfModelId: params.modelId };
1351
1327
  }
1352
1328
  return null;
@@ -2249,7 +2225,7 @@ var templates = {
2249
2225
  "basicAudio": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n data,\n model: "{{ model.id }}",\n provider: "{{ provider }}",\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(output);',
2250
2226
  "basicImage": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n data,\n model: "{{ model.id }}",\n provider: "{{ provider }}",\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(output);',
2251
2227
  "conversational": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nconst chatCompletion = await client.chatCompletion({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nconsole.log(chatCompletion.choices[0].message);',
2252
- "conversationalStream": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nlet out = "";\n\nconst stream = await client.chatCompletionStream({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nfor await (const chunk of stream) {\n if (chunk.choices && chunk.choices.length > 0) {\n const newContent = chunk.choices[0].delta.content;\n out += newContent;\n console.log(newContent);\n } \n}',
2228
+ "conversationalStream": 'import { InferenceClient } from "@huggingface/inference";\n\nconst client = new InferenceClient("{{ accessToken }}");\n\nlet out = "";\n\nconst stream = client.chatCompletionStream({\n provider: "{{ provider }}",\n model: "{{ model.id }}",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: "{{ billTo }}",\n}{% endif %});\n\nfor await (const chunk of stream) {\n if (chunk.choices && chunk.choices.length > 0) {\n const newContent = chunk.choices[0].delta.content;\n out += newContent;\n console.log(newContent);\n }\n}',
2253
2229
  "textToImage": `import { InferenceClient } from "@huggingface/inference";
2254
2230
 
2255
2231
  const client = new InferenceClient("{{ accessToken }}");
package/dist/src/lib/getInferenceProviderMapping.d.ts CHANGED
@@ -1,7 +1,7 @@
1
1
  import type { WidgetType } from "@huggingface/tasks";
2
2
  import type { InferenceProvider, ModelId } from "../types";
3
- export declare const inferenceProviderMappingCache: Map<string, Partial<Record<"black-forest-labs" | "cerebras" | "cohere" | "fal-ai" | "featherless-ai" | "fireworks-ai" | "groq" | "hf-inference" | "hyperbolic" | "nebius" | "novita" | "nscale" | "openai" | "replicate" | "sambanova" | "together", Omit<InferenceProviderModelMapping, "hfModelId" | "adapterWeightsPath">>>>;
4
- export type InferenceProviderMapping = Partial<Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId" | "adapterWeightsPath">>>;
3
+ export declare const inferenceProviderMappingCache: Map<string, Partial<Record<"black-forest-labs" | "cerebras" | "cohere" | "fal-ai" | "featherless-ai" | "fireworks-ai" | "groq" | "hf-inference" | "hyperbolic" | "nebius" | "novita" | "nscale" | "openai" | "replicate" | "sambanova" | "together", Omit<InferenceProviderModelMapping, "hfModelId">>>>;
4
+ export type InferenceProviderMapping = Partial<Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId">>>;
5
5
  export interface InferenceProviderModelMapping {
6
6
  adapter?: string;
7
7
  adapterWeightsPath?: string;
package/dist/src/lib/getInferenceProviderMapping.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"getInferenceProviderMapping.d.ts","sourceRoot":"","sources":["../../../src/lib/getInferenceProviderMapping.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AACrD,OAAO,KAAK,EAAE,iBAAiB,EAAE,OAAO,EAAE,MAAM,UAAU,CAAC;AAM3D,eAAO,MAAM,6BAA6B,iUAA+C,CAAC;AAE1F,MAAM,MAAM,wBAAwB,GAAG,OAAO,CAC7C,MAAM,CAAC,iBAAiB,EAAE,IAAI,CAAC,6BAA6B,EAAE,WAAW,GAAG,oBAAoB,CAAC,CAAC,CAClG,CAAC;AAEF,MAAM,WAAW,6BAA6B;IAC7C,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAC5B,SAAS,EAAE,OAAO,CAAC;IACnB,UAAU,EAAE,MAAM,CAAC;IACnB,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC;IAC3B,IAAI,EAAE,UAAU,CAAC;CACjB;AAED,wBAAsB,2BAA2B,CAChD,MAAM,EAAE;IACP,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,OAAO,EAAE,OAAO,CAAC;IACjB,QAAQ,EAAE,iBAAiB,CAAC;IAC5B,IAAI,EAAE,UAAU,CAAC;CACjB,EACD,OAAO,EAAE;IACR,KAAK,CAAC,EAAE,CAAC,KAAK,EAAE,WAAW,EAAE,IAAI,CAAC,EAAE,WAAW,KAAK,OAAO,CAAC,QAAQ,CAAC,CAAC;CACtE,GACC,OAAO,CAAC,6BAA6B,GAAG,IAAI,CAAC,CA+D/C"}
1
+ {"version":3,"file":"getInferenceProviderMapping.d.ts","sourceRoot":"","sources":["../../../src/lib/getInferenceProviderMapping.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AACrD,OAAO,KAAK,EAAE,iBAAiB,EAAE,OAAO,EAAE,MAAM,UAAU,CAAC;AAM3D,eAAO,MAAM,6BAA6B,0SAA+C,CAAC;AAE1F,MAAM,MAAM,wBAAwB,GAAG,OAAO,CAC7C,MAAM,CAAC,iBAAiB,EAAE,IAAI,CAAC,6BAA6B,EAAE,WAAW,CAAC,CAAC,CAC3E,CAAC;AAEF,MAAM,WAAW,6BAA6B;IAC7C,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAC5B,SAAS,EAAE,OAAO,CAAC;IACnB,UAAU,EAAE,MAAM,CAAC;IACnB,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC;IAC3B,IAAI,EAAE,UAAU,CAAC;CACjB;AAED,wBAAsB,2BAA2B,CAChD,MAAM,EAAE;IACP,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,OAAO,EAAE,OAAO,CAAC;IACjB,QAAQ,EAAE,iBAAiB,CAAC;IAC5B,IAAI,EAAE,UAAU,CAAC;CACjB,EACD,OAAO,EAAE;IACR,KAAK,CAAC,EAAE,CAAC,KAAK,EAAE,WAAW,EAAE,IAAI,CAAC,EAAE,WAAW,KAAK,OAAO,CAAC,QAAQ,CAAC,CAAC;CACtE,GACC,OAAO,CAAC,6BAA6B,GAAG,IAAI,CAAC,CA+C/C"}
package/dist/src/providers/fal-ai.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"fal-ai.d.ts","sourceRoot":"","sources":["../../../src/providers/fal-ai.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AACH,OAAO,KAAK,EAAE,gCAAgC,EAAE,MAAM,oBAAoB,CAAC;AAG3E,OAAO,KAAK,EAAE,UAAU,EAAE,YAAY,EAAW,SAAS,EAAE,MAAM,UAAU,CAAC;AAG7E,OAAO,EACN,KAAK,oCAAoC,EACzC,kBAAkB,EAClB,KAAK,qBAAqB,EAC1B,KAAK,qBAAqB,EAC1B,MAAM,kBAAkB,CAAC;AAG1B,MAAM,WAAW,gBAAgB;IAChC,UAAU,EAAE,MAAM,CAAC;IACnB,MAAM,EAAE,MAAM,CAAC;IACf,YAAY,EAAE,MAAM,CAAC;CACrB;AAED,UAAU,sBAAsB;IAC/B,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;KACZ,CAAC,CAAC;CACH;AAYD,eAAO,MAAM,2BAA2B,UAA0D,CAAC;AAEnG,uBAAe,SAAU,SAAQ,kBAAkB;gBACtC,GAAG,CAAC,EAAE,MAAM;IAIxB,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAG3D,SAAS,CAAC,MAAM,EAAE,SAAS,GAAG,MAAM;IAG3B,cAAc,CAAC,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,OAAO,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC;CAUtF;AAMD,qBAAa,oBAAqB,SAAQ,SAAU,YAAW,qBAAqB;IAC1E,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAiCrD,WAAW,CAAC,QAAQ,EAAE,sBAAsB,EAAE,UAAU,CAAC,EAAE,KAAK,GAAG,MAAM,GAAG,OAAO,CAAC,MAAM,GAAG,IAAI,CAAC;CAkBjH;AAED,qBAAa,oBAAqB,SAAQ,SAAU,YAAW,qBAAqB;;IAI1E,SAAS,CAAC,MAAM,EAAE,SAAS,GAAG,MAAM;IAMpC,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAQrD,WAAW,CACzB,QAAQ,EAAE,gBAAgB,EAC1B,GAAG,CAAC,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,GAC9B,OAAO,CAAC,IAAI,CAAC;CA8DhB;AAED,qBAAa,mCAAoC,SAAQ,SAAU,YAAW,oCAAoC;IACxG,cAAc,CAAC,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,OAAO,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC;IAKvE,WAAW,CAAC,QAAQ,EAAE,OAAO,GAAG,OAAO,CAAC,gCAAgC,CAAC;CASxF;AAED,qBAAa,qBAAsB,SAAQ,SAAS;IAC1C,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAQrD,WAAW,CAAC,QAAQ,EAAE,OAAO,GAAG,OAAO,CAAC,IAAI,CAAC;CAqB5D"}
1
+ {"version":3,"file":"fal-ai.d.ts","sourceRoot":"","sources":["../../../src/providers/fal-ai.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AACH,OAAO,KAAK,EAAE,gCAAgC,EAAE,MAAM,oBAAoB,CAAC;AAG3E,OAAO,KAAK,EAAE,UAAU,EAAE,YAAY,EAAW,SAAS,EAAE,MAAM,UAAU,CAAC;AAG7E,OAAO,EACN,KAAK,oCAAoC,EACzC,kBAAkB,EAClB,KAAK,qBAAqB,EAC1B,KAAK,qBAAqB,EAC1B,MAAM,kBAAkB,CAAC;AAG1B,MAAM,WAAW,gBAAgB;IAChC,UAAU,EAAE,MAAM,CAAC;IACnB,MAAM,EAAE,MAAM,CAAC;IACf,YAAY,EAAE,MAAM,CAAC;CACrB;AAED,UAAU,sBAAsB;IAC/B,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;KACZ,CAAC,CAAC;CACH;AAYD,eAAO,MAAM,2BAA2B,UAA0D,CAAC;AAEnG,uBAAe,SAAU,SAAQ,kBAAkB;gBACtC,GAAG,CAAC,EAAE,MAAM;IAIxB,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAG3D,SAAS,CAAC,MAAM,EAAE,SAAS,GAAG,MAAM;IAG3B,cAAc,CAAC,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,OAAO,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC;CAUtF;AAMD,qBAAa,oBAAqB,SAAQ,SAAU,YAAW,qBAAqB;IAC1E,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAuBrD,WAAW,CAAC,QAAQ,EAAE,sBAAsB,EAAE,UAAU,CAAC,EAAE,KAAK,GAAG,MAAM,GAAG,OAAO,CAAC,MAAM,GAAG,IAAI,CAAC;CAkBjH;AAED,qBAAa,oBAAqB,SAAQ,SAAU,YAAW,qBAAqB;;IAI1E,SAAS,CAAC,MAAM,EAAE,SAAS,GAAG,MAAM;IAMpC,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAQrD,WAAW,CACzB,QAAQ,EAAE,gBAAgB,EAC1B,GAAG,CAAC,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,GAC9B,OAAO,CAAC,IAAI,CAAC;CA8DhB;AAED,qBAAa,mCAAoC,SAAQ,SAAU,YAAW,oCAAoC;IACxG,cAAc,CAAC,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,OAAO,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC;IAKvE,WAAW,CAAC,QAAQ,EAAE,OAAO,GAAG,OAAO,CAAC,gCAAgC,CAAC;CASxF;AAED,qBAAa,qBAAsB,SAAQ,SAAS;IAC1C,cAAc,CAAC,MAAM,EAAE,UAAU,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAQrD,WAAW,CAAC,QAAQ,EAAE,OAAO,GAAG,OAAO,CAAC,IAAI,CAAC;CAqB5D"}
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@huggingface/inference",
3
- "version": "3.9.0",
3
+ "version": "3.9.2",
4
4
  "packageManager": "pnpm@8.10.5",
5
5
  "license": "MIT",
6
6
  "author": "Hugging Face and Tim Mikeladze <tim.mikeladze@gmail.com>",
@@ -40,8 +40,8 @@
40
40
  },
41
41
  "type": "module",
42
42
  "dependencies": {
43
- "@huggingface/jinja": "^0.3.4",
44
- "@huggingface/tasks": "^0.18.10"
43
+ "@huggingface/tasks": "^0.18.11",
44
+ "@huggingface/jinja": "^0.3.4"
45
45
  },
46
46
  "devDependencies": {
47
47
  "@types/node": "18.13.0"
package/src/lib/getInferenceProviderMapping.ts CHANGED
@@ -8,7 +8,7 @@ import { typedInclude } from "../utils/typedInclude";
8
8
  export const inferenceProviderMappingCache = new Map<ModelId, InferenceProviderMapping>();
9
9
 
10
10
  export type InferenceProviderMapping = Partial<
11
- Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId" | "adapterWeightsPath">>
11
+ Record<InferenceProvider, Omit<InferenceProviderModelMapping, "hfModelId">>
12
12
  >;
13
13
 
14
14
  export interface InferenceProviderModelMapping {
@@ -74,22 +74,6 @@ export async function getInferenceProviderMapping(
74
74
  `Model ${params.modelId} is in staging mode for provider ${params.provider}. Meant for test purposes only.`
75
75
  );
76
76
  }
77
- if (providerMapping.adapter === "lora") {
78
- const treeResp = await (options?.fetch ?? fetch)(`${HF_HUB_URL}/api/models/${params.modelId}/tree/main`);
79
- if (!treeResp.ok) {
80
- throw new Error(`Unable to fetch the model tree for ${params.modelId}.`);
81
- }
82
- const tree: Array<{ type: "file" | "directory"; path: string }> = await treeResp.json();
83
- const adapterWeightsPath = tree.find(({ type, path }) => type === "file" && path.endsWith(".safetensors"))?.path;
84
- if (!adapterWeightsPath) {
85
- throw new Error(`No .safetensors file found in the model tree for ${params.modelId}.`);
86
- }
87
- return {
88
- ...providerMapping,
89
- hfModelId: params.modelId,
90
- adapterWeightsPath,
91
- };
92
- }
93
77
  return { ...providerMapping, hfModelId: params.modelId };
94
78
  }
95
79
  return null;
package/src/providers/fal-ai.ts CHANGED
@@ -86,16 +86,6 @@ export class FalAITextToImageTask extends FalAITask implements TextToImageTaskHe
86
86
  ...(params.args.parameters as Record<string, unknown>),
87
87
  sync_mode: true,
88
88
  prompt: params.args.inputs,
89
- ...(params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath
90
- ? {
91
- loras: [
92
- {
93
- path: buildLoraPath(params.mapping.hfModelId, params.mapping.adapterWeightsPath),
94
- scale: 1,
95
- },
96
- ],
97
- }
98
- : undefined),
99
89
  };
100
90
 
101
91
  if (params.mapping?.adapter === "lora" && params.mapping.adapterWeightsPath) {
package/src/snippets/templates.exported.ts CHANGED
@@ -14,7 +14,7 @@ export const templates: Record<string, Record<string, Record<string, string>>> =
14
14
  "basicAudio": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n\tdata,\n\tmodel: \"{{ model.id }}\",\n\tprovider: \"{{ provider }}\",\n}{% if billTo %}, {\n\tbillTo: \"{{ billTo }}\",\n}{% endif %});\n\nconsole.log(output);",
15
15
  "basicImage": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nconst data = fs.readFileSync({{inputs.asObj.inputs}});\n\nconst output = await client.{{ methodName }}({\n\tdata,\n\tmodel: \"{{ model.id }}\",\n\tprovider: \"{{ provider }}\",\n}{% if billTo %}, {\n\tbillTo: \"{{ billTo }}\",\n}{% endif %});\n\nconsole.log(output);",
16
16
  "conversational": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nconst chatCompletion = await client.chatCompletion({\n provider: \"{{ provider }}\",\n model: \"{{ model.id }}\",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: \"{{ billTo }}\",\n}{% endif %});\n\nconsole.log(chatCompletion.choices[0].message);",
17
- "conversationalStream": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nlet out = \"\";\n\nconst stream = await client.chatCompletionStream({\n provider: \"{{ provider }}\",\n model: \"{{ model.id }}\",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: \"{{ billTo }}\",\n}{% endif %});\n\nfor await (const chunk of stream) {\n\tif (chunk.choices && chunk.choices.length > 0) {\n\t\tconst newContent = chunk.choices[0].delta.content;\n\t\tout += newContent;\n\t\tconsole.log(newContent);\n\t} \n}",
17
+ "conversationalStream": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nlet out = \"\";\n\nconst stream = client.chatCompletionStream({\n provider: \"{{ provider }}\",\n model: \"{{ model.id }}\",\n{{ inputs.asTsString }}\n}{% if billTo %}, {\n billTo: \"{{ billTo }}\",\n}{% endif %});\n\nfor await (const chunk of stream) {\n\tif (chunk.choices && chunk.choices.length > 0) {\n\t\tconst newContent = chunk.choices[0].delta.content;\n\t\tout += newContent;\n\t\tconsole.log(newContent);\n\t}\n}",
18
18
  "textToImage": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nconst image = await client.textToImage({\n provider: \"{{ provider }}\",\n model: \"{{ model.id }}\",\n\tinputs: {{ inputs.asObj.inputs }},\n\tparameters: { num_inference_steps: 5 },\n}{% if billTo %}, {\n billTo: \"{{ billTo }}\",\n}{% endif %});\n/// Use the generated image (it's a Blob)",
19
19
  "textToVideo": "import { InferenceClient } from \"@huggingface/inference\";\n\nconst client = new InferenceClient(\"{{ accessToken }}\");\n\nconst image = await client.textToVideo({\n provider: \"{{ provider }}\",\n model: \"{{ model.id }}\",\n\tinputs: {{ inputs.asObj.inputs }},\n}{% if billTo %}, {\n billTo: \"{{ billTo }}\",\n}{% endif %});\n// Use the generated video (it's a Blob)"
20
20
  },