Skip to content

feat(models): add vision input capabilities - #43

Merged
patlux merged 1 commit into
mainfrom
fix/issue-42-model-capabilities
Aug 10, 2026
Merged

feat(models): add vision input capabilities#43
patlux merged 1 commit into
mainfrom
fix/issue-42-model-capabilities

Conversation

@patlux

@patlux patlux commented Aug 10, 2026

Copy link
Copy Markdown
Owner

Summary

  • sync model image input capabilities with the bundled command-code@1.15.1 catalog
  • advertise image input only for upstream-confirmed vision models, including GPT-5.6 Luna and Muse Spark 1.2
  • serialize user and tool-result images using Command Code's current data-URL wire format
  • keep unknown and text-only models fail-closed before network access
  • update the Command Code client version header and image-input documentation

Closes #42

Verification

  • npm test
  • npm run format:check
  • git diff --check

npm test passed; the optional OMP compatibility smoke test was skipped because omp is not installed on PATH.

@patlux patlux self-assigned this Aug 10, 2026
@github-actions

Copy link
Copy Markdown

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

@github-actions

Copy link
Copy Markdown

Memory benchmark

Compared base 4b6e1cd with PR head bd056d0 on the same GitHub-hosted darwin runner. Lower values are better.

Metric Base PR PR − Base Change
Stable RSS 167.8 MiB ± 0.6 MiB 168.5 MiB ± 0.8 MiB +0.5 MiB ± 1.0 MiB +0.3%
Physical footprint 110.6 MiB ± 0.4 MiB 111.3 MiB ± 0.6 MiB +0.1 MiB ± 0.2 MiB +0.1%
Physical peak 127.3 MiB ± 0.8 MiB 128.6 MiB ± 1.0 MiB +0.5 MiB ± 1.1 MiB +0.4%

Interpretation: No metric shows a clear Base-to-PR difference beyond its measured run-to-run variation.

Extension overhead above pi baseline

Metric Base overhead PR overhead Difference
Stable RSS +6.1 MiB ± 1.1 MiB +6.8 MiB ± 1.5 MiB +0.5 MiB ± 1.0 MiB
Physical footprint +4.0 MiB ± 0.9 MiB +4.8 MiB ± 1.2 MiB +0.1 MiB ± 0.2 MiB
Physical peak +3.7 MiB ± 1.7 MiB +4.7 MiB ± 1.8 MiB +0.5 MiB ± 1.1 MiB

Values are medians of 6 alternating, paired runs. The value after ± is the median absolute deviation (MAD). Each process was sampled 12 times after a 2500 ms warm-up.

Environment: pi 0.82.1, Bun 1.3.11, darwin arm64. RSS comes from ps; physical footprint and peak come from macOS footprint.

This is a comparative signal, not a pass/fail threshold. GitHub-hosted runner noise can affect absolute values.

@patlux

patlux commented Aug 10, 2026

Copy link
Copy Markdown
Owner Author

Attachments

Authenticated pi vision E2E

Screenshot_2026-08-11_00-26-18@2x.png
Screenshot_2026-08-11_00-26-18@2x.png

@patlux
patlux merged commit a367aed into main Aug 10, 2026
12 checks passed
@patlux
patlux deleted the fix/issue-42-model-capabilities branch August 10, 2026 22:28
@patlux

patlux commented Aug 10, 2026

Copy link
Copy Markdown
Owner Author

Shipped in pi-commandcode-provider@0.5.1 / GitHub release v0.5.1.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Vision not supported?

1 participant