ponytail

132 Commits 30 Branches 14 Tags

3 Commits

Include renames

Author	SHA1	Message	Date
Tanmay Garg	08f0daffbb	fix(benchmark): scheme validation to ollama-url (closes #166 ) (#274 ) Add urllib.parse.urlparse to benchmark-local.py and validate that the provided --ollama-url uses either the http or https scheme. Fix arbitrary URI handling where the script could previously access local files (file://) or other unsupported protocols. If the scheme is invalid, parser.error is called to exit cleanly with a clear message.	2026-06-24 00:44:47 +02:00
DietrichGebertandClaude Opus 4.8	2e6a93765a	fix(benchmarks): count unfenced code, ASCII-safe output, refresh llama3.2 results (#67 ) Fixes the local benchmark LOC counter (counted only fenced code, scored bare output 0), makes summary output ASCII-safe (a Unicode arrow crashed the script on Windows cp1252), gitignores generated artifacts, and refreshes the llama3.2 writeup with n=5 data showing the LOC effect is within the noise floor. Follow-up to #63. Verified live. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>	2026-06-15 16:22:32 +02:00
Mandavilli Vijay	386f95734a	benchmarks: add local model support and Node version note (#63 ) Adds benchmarks/benchmark-local.py (Ollama-based local runner), a results writeup, and a Node version note. Thanks @mandavillivijay.	2026-06-15 15:27:11 +02:00

Author

SHA1

Message

Date

Tanmay Garg

08f0daffbb

fix(benchmark): scheme validation to ollama-url (closes #166 ) (#274 )

Add urllib.parse.urlparse to benchmark-local.py and validate that the
provided --ollama-url uses either the http or https scheme.

Fix arbitrary URI handling where the script could previously access
local files (file://) or other unsupported protocols. If the scheme
is invalid, parser.error is called to exit cleanly with a clear message.

2026-06-24 00:44:47 +02:00

DietrichGebertandClaude Opus 4.8

2e6a93765a

fix(benchmarks): count unfenced code, ASCII-safe output, refresh llama3.2 results (#67 )

Fixes the local benchmark LOC counter (counted only fenced code, scored bare output 0), makes summary output ASCII-safe (a Unicode arrow crashed the script on Windows cp1252), gitignores generated artifacts, and refreshes the llama3.2 writeup with n=5 data showing the LOC effect is within the noise floor. Follow-up to #63. Verified live.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-15 16:22:32 +02:00

Mandavilli Vijay

386f95734a

benchmarks: add local model support and Node version note (#63 )

Adds benchmarks/benchmark-local.py (Ollama-based local runner), a results writeup, and a Node version note. Thanks @mandavillivijay.

2026-06-15 15:27:11 +02:00