@huggingface/tasks 0.20.24 → 0.20.25

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 (8) hide show
  1. package/dist/commonjs/eval.d.ts +10 -0
  2. package/dist/commonjs/eval.d.ts.map +1 -1
  3. package/dist/commonjs/eval.js +10 -0
  4. package/dist/esm/eval.d.ts +10 -0
  5. package/dist/esm/eval.d.ts.map +1 -1
  6. package/dist/esm/eval.js +10 -0
  7. package/package.json +1 -1
  8. package/src/eval.ts +12 -0
package/dist/commonjs/eval.d.ts CHANGED
@@ -92,5 +92,15 @@ export declare const EVALUATION_FRAMEWORKS: {
92
92
  readonly description: "CLAW-Eval is an evaluation framework for assessing LLMs as autonomous agents across 300 human-verified tasks covering communication, finance, and productivity domains.";
93
93
  readonly url: "https://github.com/claw-eval/claw-eval";
94
94
  };
95
+ readonly pbench: {
96
+ readonly name: "pbench";
97
+ readonly description: "PBench is a multi-level referring expression segmentation benchmark for evaluating vision-language perception across a structured hierarchy of skills.";
98
+ readonly url: "https://github.com/tiiuae/Falcon-Perception";
99
+ };
100
+ readonly wildclawbench: {
101
+ readonly name: "wildclawbench";
102
+ readonly description: "WildClawBench is an in-the-wild benchmark for evaluating AI agents in the OpenClaw environment across 60 hand-built, end-to-end tasks spanning productivity, code intelligence, social interaction, search, creative synthesis, and safety domains.";
103
+ readonly url: "https://github.com/InternLM/WildClawBench";
104
+ };
95
105
  };
96
106
  //# sourceMappingURL=eval.d.ts.map
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;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAoGxB,CAAC"}
1
+ {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAgHxB,CAAC"}
package/dist/commonjs/eval.js CHANGED
@@ -95,4 +95,14 @@ exports.EVALUATION_FRAMEWORKS = {
95
95
  description: "CLAW-Eval is an evaluation framework for assessing LLMs as autonomous agents across 300 human-verified tasks covering communication, finance, and productivity domains.",
96
96
  url: "https://github.com/claw-eval/claw-eval",
97
97
  },
98
+ pbench: {
99
+ name: "pbench",
100
+ description: "PBench is a multi-level referring expression segmentation benchmark for evaluating vision-language perception across a structured hierarchy of skills.",
101
+ url: "https://github.com/tiiuae/Falcon-Perception",
102
+ },
103
+ wildclawbench: {
104
+ name: "wildclawbench",
105
+ description: "WildClawBench is an in-the-wild benchmark for evaluating AI agents in the OpenClaw environment across 60 hand-built, end-to-end tasks spanning productivity, code intelligence, social interaction, search, creative synthesis, and safety domains.",
106
+ url: "https://github.com/InternLM/WildClawBench",
107
+ },
98
108
  };
package/dist/esm/eval.d.ts CHANGED
@@ -92,5 +92,15 @@ export declare const EVALUATION_FRAMEWORKS: {
92
92
  readonly description: "CLAW-Eval is an evaluation framework for assessing LLMs as autonomous agents across 300 human-verified tasks covering communication, finance, and productivity domains.";
93
93
  readonly url: "https://github.com/claw-eval/claw-eval";
94
94
  };
95
+ readonly pbench: {
96
+ readonly name: "pbench";
97
+ readonly description: "PBench is a multi-level referring expression segmentation benchmark for evaluating vision-language perception across a structured hierarchy of skills.";
98
+ readonly url: "https://github.com/tiiuae/Falcon-Perception";
99
+ };
100
+ readonly wildclawbench: {
101
+ readonly name: "wildclawbench";
102
+ readonly description: "WildClawBench is an in-the-wild benchmark for evaluating AI agents in the OpenClaw environment across 60 hand-built, end-to-end tasks spanning productivity, code intelligence, social interaction, search, creative synthesis, and safety domains.";
103
+ readonly url: "https://github.com/InternLM/WildClawBench";
104
+ };
95
105
  };
96
106
  //# sourceMappingURL=eval.d.ts.map
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;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAoGxB,CAAC"}
1
+ {"version":3,"file":"eval.d.ts","sourceRoot":"","sources":["../../src/eval.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAgHxB,CAAC"}
package/dist/esm/eval.js CHANGED
@@ -92,4 +92,14 @@ export const EVALUATION_FRAMEWORKS = {
92
92
  description: "CLAW-Eval is an evaluation framework for assessing LLMs as autonomous agents across 300 human-verified tasks covering communication, finance, and productivity domains.",
93
93
  url: "https://github.com/claw-eval/claw-eval",
94
94
  },
95
+ pbench: {
96
+ name: "pbench",
97
+ description: "PBench is a multi-level referring expression segmentation benchmark for evaluating vision-language perception across a structured hierarchy of skills.",
98
+ url: "https://github.com/tiiuae/Falcon-Perception",
99
+ },
100
+ wildclawbench: {
101
+ name: "wildclawbench",
102
+ description: "WildClawBench is an in-the-wild benchmark for evaluating AI agents in the OpenClaw environment across 60 hand-built, end-to-end tasks spanning productivity, code intelligence, social interaction, search, creative synthesis, and safety domains.",
103
+ url: "https://github.com/InternLM/WildClawBench",
104
+ },
95
105
  };
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@huggingface/tasks",
3
- "version": "0.20.24",
3
+ "version": "0.20.25",
4
4
  "description": "List of ML tasks for huggingface.co/tasks",
5
5
  "keywords": [
6
6
  "hub",
package/src/eval.ts CHANGED
@@ -101,4 +101,16 @@ export const EVALUATION_FRAMEWORKS = {
101
101
  "CLAW-Eval is an evaluation framework for assessing LLMs as autonomous agents across 300 human-verified tasks covering communication, finance, and productivity domains.",
102
102
  url: "https://github.com/claw-eval/claw-eval",
103
103
  },
104
+ pbench: {
105
+ name: "pbench",
106
+ description:
107
+ "PBench is a multi-level referring expression segmentation benchmark for evaluating vision-language perception across a structured hierarchy of skills.",
108
+ url: "https://github.com/tiiuae/Falcon-Perception",
109
+ },
110
+ wildclawbench: {
111
+ name: "wildclawbench",
112
+ description:
113
+ "WildClawBench is an in-the-wild benchmark for evaluating AI agents in the OpenClaw environment across 60 hand-built, end-to-end tasks spanning productivity, code intelligence, social interaction, search, creative synthesis, and safety domains.",
114
+ url: "https://github.com/InternLM/WildClawBench",
115
+ },
104
116
  } as const;