@huggingface/tasks 0.21.16 → 0.21.17

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 (13) hide show
  1. package/dist/commonjs/eval.d.ts +5 -0
  2. package/dist/commonjs/eval.d.ts.map +1 -1
  3. package/dist/commonjs/eval.js +5 -0
  4. package/dist/commonjs/hardware-nvidia.d.ts.map +1 -1
  5. package/dist/commonjs/hardware-nvidia.js +8 -0
  6. package/dist/esm/eval.d.ts +5 -0
  7. package/dist/esm/eval.d.ts.map +1 -1
  8. package/dist/esm/eval.js +5 -0
  9. package/dist/esm/hardware-nvidia.d.ts.map +1 -1
  10. package/dist/esm/hardware-nvidia.js +8 -0
  11. package/package.json +1 -1
  12. package/src/eval.ts +6 -0
  13. package/src/hardware-nvidia.ts +8 -0
package/dist/commonjs/eval.d.ts CHANGED
@@ -47,6 +47,11 @@ export declare const EVALUATION_FRAMEWORKS: {
47
47
  readonly description: "Archipelago is a system for running and evaluating AI agents against MCP applications.";
48
48
  readonly url: "https://github.com/Mercor-Intelligence/archipelago";
49
49
  };
50
+ readonly benchflow: {
51
+ readonly name: "benchflow";
52
+ readonly description: "BenchFlow is an evaluation framework for AI agents on professional, skill-aware workflows. It powers SkillsBench and runs containerized agent trials with paired with-skills / without-skills configurations.";
53
+ readonly url: "https://github.com/benchflow-ai/benchflow";
54
+ };
50
55
  readonly "apex-evals": {
51
56
  readonly name: "apex-evals";
52
57
  readonly description: "APEX Evals is a benchmark suite and evaluation harness for evaluating large language models.";
package/dist/commonjs/eval.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAoJxB,CAAC"}
1
+ {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA0JxB,CAAC"}
package/dist/commonjs/eval.js CHANGED
@@ -50,6 +50,11 @@ exports.EVALUATION_FRAMEWORKS = {
50
50
  description: "Archipelago is a system for running and evaluating AI agents against MCP applications.",
51
51
  url: "https://github.com/Mercor-Intelligence/archipelago",
52
52
  },
53
+ benchflow: {
54
+ name: "benchflow",
55
+ description: "BenchFlow is an evaluation framework for AI agents on professional, skill-aware workflows. It powers SkillsBench and runs containerized agent trials with paired with-skills / without-skills configurations.",
56
+ url: "https://github.com/benchflow-ai/benchflow",
57
+ },
53
58
  "apex-evals": {
54
59
  name: "apex-evals",
55
60
  description: "APEX Evals is a benchmark suite and evaluation harness for evaluating large language models.",
package/dist/commonjs/hardware-nvidia.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"hardware-nvidia.d.ts","sourceRoot":"","sources":["../../src/hardware-nvidia.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,eAAe,CAAC;AAElD,MAAM,WAAW,kBAAmB,SAAQ,YAAY;IACvD;;;;OAIG;IACH,iBAAiB,EAAE,MAAM,CAAC;CAC1B;AAED,oBAAY,yBAAyB;IACpC,eAAe,OAAO;IACtB,aAAa,KAAO;IACpB,SAAS,KAAO;IAChB,MAAM,IAAM;IACZ,YAAY,MAAM;IAClB,IAAI,MAAM;IACV,UAAU,MAAM;IAChB,MAAM,IAAM;IACZ,MAAM,MAAM;IACZ,MAAM,MAAM;IACZ,KAAK,IAAM;IACX,YAAY,MAAM;IAClB,MAAM,MAAM;IACZ,iBAAiB,IAAM;IACvB,OAAO,MAAM;CACb;AAED,eAAO,MAAM,WAAW,EAAE,MAAM,CAAC,MAAM,EAAE,kBAAkB,CAy8B1D,CAAC"}
1
+ {"version":3,"file":"hardware-nvidia.d.ts","sourceRoot":"","sources":["../../src/hardware-nvidia.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,eAAe,CAAC;AAElD,MAAM,WAAW,kBAAmB,SAAQ,YAAY;IACvD;;;;OAIG;IACH,iBAAiB,EAAE,MAAM,CAAC;CAC1B;AAED,oBAAY,yBAAyB;IACpC,eAAe,OAAO;IACtB,aAAa,KAAO;IACpB,SAAS,KAAO;IAChB,MAAM,IAAM;IACZ,YAAY,MAAM;IAClB,IAAI,MAAM;IACV,UAAU,MAAM;IAChB,MAAM,IAAM;IACZ,MAAM,MAAM;IACZ,MAAM,MAAM;IACZ,KAAK,IAAM;IACX,YAAY,MAAM;IAClB,MAAM,MAAM;IACZ,iBAAiB,IAAM;IACvB,OAAO,MAAM;CACb;AAED,eAAO,MAAM,WAAW,EAAE,MAAM,CAAC,MAAM,EAAE,kBAAkB,CAi9B1D,CAAC"}
package/dist/commonjs/hardware-nvidia.js CHANGED
@@ -212,6 +212,14 @@ exports.NVIDIA_SKUS = {
212
212
  power: 70,
213
213
  releaseYear: 2023,
214
214
  },
215
+ "RTX 3500 Ada Mobile": {
216
+ tflops: 15.8,
217
+ memory: [12],
218
+ computeCapability: 8.9,
219
+ msrp: 1_500,
220
+ power: 150,
221
+ releaseYear: 2023,
222
+ },
215
223
  "RTX 2000 Ada": {
216
224
  tflops: 12.0,
217
225
  memory: [16],
package/dist/esm/eval.d.ts CHANGED
@@ -47,6 +47,11 @@ export declare const EVALUATION_FRAMEWORKS: {
47
47
  readonly description: "Archipelago is a system for running and evaluating AI agents against MCP applications.";
48
48
  readonly url: "https://github.com/Mercor-Intelligence/archipelago";
49
49
  };
50
+ readonly benchflow: {
51
+ readonly name: "benchflow";
52
+ readonly description: "BenchFlow is an evaluation framework for AI agents on professional, skill-aware workflows. It powers SkillsBench and runs containerized agent trials with paired with-skills / without-skills configurations.";
53
+ readonly url: "https://github.com/benchflow-ai/benchflow";
54
+ };
50
55
  readonly "apex-evals": {
51
56
  readonly name: "apex-evals";
52
57
  readonly description: "APEX Evals is a benchmark suite and evaluation harness for evaluating large language models.";
package/dist/esm/eval.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAoJxB,CAAC"}
1
+ {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA0JxB,CAAC"}
package/dist/esm/eval.js CHANGED
@@ -47,6 +47,11 @@ export const EVALUATION_FRAMEWORKS = {
47
47
  description: "Archipelago is a system for running and evaluating AI agents against MCP applications.",
48
48
  url: "https://github.com/Mercor-Intelligence/archipelago",
49
49
  },
50
+ benchflow: {
51
+ name: "benchflow",
52
+ description: "BenchFlow is an evaluation framework for AI agents on professional, skill-aware workflows. It powers SkillsBench and runs containerized agent trials with paired with-skills / without-skills configurations.",
53
+ url: "https://github.com/benchflow-ai/benchflow",
54
+ },
50
55
  "apex-evals": {
51
56
  name: "apex-evals",
52
57
  description: "APEX Evals is a benchmark suite and evaluation harness for evaluating large language models.",
package/dist/esm/hardware-nvidia.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"hardware-nvidia.d.ts","sourceRoot":"","sources":["../../src/hardware-nvidia.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,eAAe,CAAC;AAElD,MAAM,WAAW,kBAAmB,SAAQ,YAAY;IACvD;;;;OAIG;IACH,iBAAiB,EAAE,MAAM,CAAC;CAC1B;AAED,oBAAY,yBAAyB;IACpC,eAAe,OAAO;IACtB,aAAa,KAAO;IACpB,SAAS,KAAO;IAChB,MAAM,IAAM;IACZ,YAAY,MAAM;IAClB,IAAI,MAAM;IACV,UAAU,MAAM;IAChB,MAAM,IAAM;IACZ,MAAM,MAAM;IACZ,MAAM,MAAM;IACZ,KAAK,IAAM;IACX,YAAY,MAAM;IAClB,MAAM,MAAM;IACZ,iBAAiB,IAAM;IACvB,OAAO,MAAM;CACb;AAED,eAAO,MAAM,WAAW,EAAE,MAAM,CAAC,MAAM,EAAE,kBAAkB,CAy8B1D,CAAC"}
1
+ {"version":3,"file":"hardware-nvidia.d.ts","sourceRoot":"","sources":["../../src/hardware-nvidia.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,eAAe,CAAC;AAElD,MAAM,WAAW,kBAAmB,SAAQ,YAAY;IACvD;;;;OAIG;IACH,iBAAiB,EAAE,MAAM,CAAC;CAC1B;AAED,oBAAY,yBAAyB;IACpC,eAAe,OAAO;IACtB,aAAa,KAAO;IACpB,SAAS,KAAO;IAChB,MAAM,IAAM;IACZ,YAAY,MAAM;IAClB,IAAI,MAAM;IACV,UAAU,MAAM;IAChB,MAAM,IAAM;IACZ,MAAM,MAAM;IACZ,MAAM,MAAM;IACZ,KAAK,IAAM;IACX,YAAY,MAAM;IAClB,MAAM,MAAM;IACZ,iBAAiB,IAAM;IACvB,OAAO,MAAM;CACb;AAED,eAAO,MAAM,WAAW,EAAE,MAAM,CAAC,MAAM,EAAE,kBAAkB,CAi9B1D,CAAC"}
package/dist/esm/hardware-nvidia.js CHANGED
@@ -209,6 +209,14 @@ export const NVIDIA_SKUS = {
209
209
  power: 70,
210
210
  releaseYear: 2023,
211
211
  },
212
+ "RTX 3500 Ada Mobile": {
213
+ tflops: 15.8,
214
+ memory: [12],
215
+ computeCapability: 8.9,
216
+ msrp: 1_500,
217
+ power: 150,
218
+ releaseYear: 2023,
219
+ },
212
220
  "RTX 2000 Ada": {
213
221
  tflops: 12.0,
214
222
  memory: [16],
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@huggingface/tasks",
3
- "version": "0.21.16",
3
+ "version": "0.21.17",
4
4
  "description": "List of ML tasks for huggingface.co/tasks",
5
5
  "keywords": [
6
6
  "hub",
package/src/eval.ts CHANGED
@@ -50,6 +50,12 @@ export const EVALUATION_FRAMEWORKS = {
50
50
  description: "Archipelago is a system for running and evaluating AI agents against MCP applications.",
51
51
  url: "https://github.com/Mercor-Intelligence/archipelago",
52
52
  },
53
+ benchflow: {
54
+ name: "benchflow",
55
+ description:
56
+ "BenchFlow is an evaluation framework for AI agents on professional, skill-aware workflows. It powers SkillsBench and runs containerized agent trials with paired with-skills / without-skills configurations.",
57
+ url: "https://github.com/benchflow-ai/benchflow",
58
+ },
53
59
  "apex-evals": {
54
60
  name: "apex-evals",
55
61
  description: "APEX Evals is a benchmark suite and evaluation harness for evaluating large language models.",
package/src/hardware-nvidia.ts CHANGED
@@ -220,6 +220,14 @@ export const NVIDIA_SKUS: Record<string, NvidiaHardwareSpec> = {
220
220
  power: 70,
221
221
  releaseYear: 2023,
222
222
  },
223
+ "RTX 3500 Ada Mobile": {
224
+ tflops: 15.8,
225
+ memory: [12],
226
+ computeCapability: 8.9,
227
+ msrp: 1_500,
228
+ power: 150,
229
+ releaseYear: 2023,
230
+ },
223
231
  "RTX 2000 Ada": {
224
232
  tflops: 12.0,
225
233
  memory: [16],