everybench 0.1.0__tar.gz
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.
|
@@ -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 +1,11 @@
|
|
|
1
|
+
# EveryBench
|
|
2
|
+
|
|
3
|
+
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).
|
|
4
|
+
|
|
5
|
+
```bash
|
|
6
|
+
uvx everybench catalog
|
|
7
|
+
uvx everybench run toy-arith -m mock/strong -m mock/weak
|
|
8
|
+
uvx everybench serve # web UI at http://127.0.0.1:8321
|
|
9
|
+
```
|
|
10
|
+
|
|
11
|
+
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 +1,26 @@
|
|
|
1
|
+
[project]
|
|
2
|
+
name = "everybench"
|
|
3
|
+
version = "0.1.0"
|
|
4
|
+
description = "EveryBench by Eva AI: run any benchmark, trace every task, compare runs with paired statistics."
|
|
5
|
+
readme = "README.md"
|
|
6
|
+
license = "Apache-2.0"
|
|
7
|
+
requires-python = ">=3.12"
|
|
8
|
+
dependencies = ["everyeval>=0.1.0"]
|
|
9
|
+
|
|
10
|
+
[project.optional-dependencies]
|
|
11
|
+
benchmarks = ["everyeval[benchmarks]>=0.1.0"]
|
|
12
|
+
harbor = ["everyeval[harbor]>=0.1.0"]
|
|
13
|
+
postgres = ["everyeval[postgres]>=0.1.0"]
|
|
14
|
+
eee = ["everyeval[eee]>=0.1.0"]
|
|
15
|
+
|
|
16
|
+
[project.scripts]
|
|
17
|
+
everybench = "everyeval.cli:app"
|
|
18
|
+
|
|
19
|
+
[project.urls]
|
|
20
|
+
Homepage = "https://evabench.dev/everybench"
|
|
21
|
+
Source = "https://github.com/trycedar0x/benchmark-tracing"
|
|
22
|
+
Issues = "https://github.com/trycedar0x/benchmark-tracing/issues"
|
|
23
|
+
|
|
24
|
+
[build-system]
|
|
25
|
+
requires = ["uv_build>=0.11.7,<0.12.0"]
|
|
26
|
+
build-backend = "uv_build"
|
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
[project]
|
|
2
|
+
name = "everybench"
|
|
3
|
+
version = "0.1.0"
|
|
4
|
+
description = "EveryBench by Eva AI: run any benchmark, trace every task, compare runs with paired statistics."
|
|
5
|
+
readme = "README.md"
|
|
6
|
+
license = "Apache-2.0"
|
|
7
|
+
requires-python = ">=3.12"
|
|
8
|
+
dependencies = ["everyeval>=0.1.0"]
|
|
9
|
+
|
|
10
|
+
[project.optional-dependencies]
|
|
11
|
+
benchmarks = ["everyeval[benchmarks]>=0.1.0"]
|
|
12
|
+
harbor = ["everyeval[harbor]>=0.1.0"]
|
|
13
|
+
postgres = ["everyeval[postgres]>=0.1.0"]
|
|
14
|
+
eee = ["everyeval[eee]>=0.1.0"]
|
|
15
|
+
|
|
16
|
+
[project.scripts]
|
|
17
|
+
everybench = "everyeval.cli:app"
|
|
18
|
+
|
|
19
|
+
[project.urls]
|
|
20
|
+
Homepage = "https://evabench.dev/everybench"
|
|
21
|
+
Source = "https://github.com/trycedar0x/benchmark-tracing"
|
|
22
|
+
Issues = "https://github.com/trycedar0x/benchmark-tracing/issues"
|
|
23
|
+
|
|
24
|
+
[build-system]
|
|
25
|
+
requires = ["uv_build>=0.11.7,<0.12.0"]
|
|
26
|
+
build-backend = "uv_build"
|