everybench 0.1.0__py3-none-any.whl

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.
everybench/__init__.py ADDED
@@ -0,0 +1,3 @@
1
+ """EveryBench. Until the code moves here, this package re-exports everyeval."""
2
+
3
+ from everyeval import * # noqa: F401,F403
@@ -0,0 +1,31 @@
1
+ Metadata-Version: 2.4
2
+ Name: everybench
3
+ Version: 0.1.0
4
+ Summary: EveryBench by Eva AI: run any benchmark, trace every task, compare runs with paired statistics.
5
+ License-Expression: Apache-2.0
6
+ Requires-Dist: everyeval>=0.1.0
7
+ Requires-Dist: everyeval[benchmarks]>=0.1.0 ; extra == 'benchmarks'
8
+ Requires-Dist: everyeval[eee]>=0.1.0 ; extra == 'eee'
9
+ Requires-Dist: everyeval[harbor]>=0.1.0 ; extra == 'harbor'
10
+ Requires-Dist: everyeval[postgres]>=0.1.0 ; extra == 'postgres'
11
+ Requires-Python: >=3.12
12
+ Project-URL: Homepage, https://evabench.dev/everybench
13
+ Project-URL: Source, https://github.com/trycedar0x/benchmark-tracing
14
+ Project-URL: Issues, https://github.com/trycedar0x/benchmark-tracing/issues
15
+ Provides-Extra: benchmarks
16
+ Provides-Extra: eee
17
+ Provides-Extra: harbor
18
+ Provides-Extra: postgres
19
+ Description-Content-Type: text/markdown
20
+
21
+ # EveryBench
22
+
23
+ Run established benchmarks against any model, see a trace for every task, and compare runs with paired statistics. Open source, local-first, no account needed. By [Eva AI](https://evabench.dev/everybench).
24
+
25
+ ```bash
26
+ uvx everybench catalog
27
+ uvx everybench run toy-arith -m mock/strong -m mock/weak
28
+ uvx everybench serve # web UI at http://127.0.0.1:8321
29
+ ```
30
+
31
+ This release installs [`everyeval`](https://pypi.org/project/everyeval/) under the new `everybench` command; the code is moving to this name. Source: [github.com/trycedar0x/benchmark-tracing](https://github.com/trycedar0x/benchmark-tracing).
@@ -0,0 +1,5 @@
1
+ everybench/__init__.py,sha256=nobPOXI9FoZCz5nZzr5luPfm7TEe9-3fzhskHF67RU0,124
2
+ everybench-0.1.0.dist-info/WHEEL,sha256=xTCBFIZIIyRSxyowOv5Wqg_jFU6KsdIffuulBB09Ph8,80
3
+ everybench-0.1.0.dist-info/entry_points.txt,sha256=HWdF0EWlwxkIuVWOx9OBB2dqc5us9d0EEsLnlnQpxNs,50
4
+ everybench-0.1.0.dist-info/METADATA,sha256=NQY49UZEM1mQnxOTsnLcJkF41s5m50MCuRl2ngdpcVg,1446
5
+ everybench-0.1.0.dist-info/RECORD,,
@@ -0,0 +1,4 @@
1
+ Wheel-Version: 1.0
2
+ Generator: uv 0.13.0
3
+ Root-Is-Purelib: true
4
+ Tag: py3-none-any
@@ -0,0 +1,3 @@
1
+ [console_scripts]
2
+ everybench = everyeval.cli:app
3
+