fix(deps): raise the tornado and pypdf floors for six new advisories

[mateo-berri]

TLDR

Problem this solves:

• osv-scan is red on every branch since 2026-09-01 20:32Z
• Three new tornado advisories flag 6.5.7 in uv.lock
• Three new pypdf advisories flag 6.15.0 in uv.lock
• All six are fixed in tornado 6.5.8 and pypdf 6.16.1

How it solves it:

• Raises the tornado constraint floor from 6.5.6 to 6.5.8
• Raises the pypdf floor in the proxy-runtime extra from 6.12.0 to 6.16.1
• Re-locks so uv.lock resolves tornado 6.5.8 and pypdf 6.16.2

User Flow

Before: anyone who scans a litellm install or opens a PR sees tornado and pypdf flagged, with CI red through no fault of their change

1. A contributor opens any PR and the osv-scan check fails with six findings from uv.lock: pypdf 6.15.0 flagged by https://osv.dev/GHSA-jp53-mhqp-8xcg (CVSS 6.9), https://osv.dev/GHSA-23w6-3w8w-8484 (CVSS 4.8) and https://osv.dev/GHSA-763m-79hh-57f2 (CVSS 4.8), and tornado 6.5.7 flagged by https://osv.dev/GHSA-mpf4-983q-p7j4 (CVSS 7.5), https://osv.dev/GHSA-8423-8fgw-73vq (CVSS 6.9) and https://osv.dev/GHSA-wwv5-g3v4-889x (CVSS 2.3)
2. An operator running a vulnerability scanner against their litellm environment gets the same six findings

After: the same scans come back clean

1. The contributor's PR runs osv-scan and it passes; no pypdf or tornado findings
2. The operator's scanner sees tornado 6.5.8 and pypdf 6.16.2, which none of the six advisories cover

Relevant issues

Closes #39217

osv-scan went red on litellm_internal_staging itself between the 19:50Z run (green) and the 20:32Z run (red) on 2026-09-01, with no lockfile change in between: the tornado advisories were published upstream that evening, the pypdf ones about an hour later, and a third tornado advisory this morning. One PR carries both bumps rather than a second uv.lock PR needing its own two approvals. Same shape as #32985 (soupsieve) and #33233 (httplib2/setuptools floors)

Supersedes #39277 and #39285 (lock-only bumps of the same two packages) and dependabot's #39191 (tornado) and #39205 (pypdf 6.16.1, which the lock already passes)

Linear ticket

Resolves LIT-6648

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have added meaningful tests (n/a: dependency floor bump, verified by the osv-scan CI check and the local scan below)
• The handful of test files covering my change pass locally, e.g. uv run pytest tests/test_litellm/<your_test_file>.py -v. Leave the suites (make test-unit-*, make test-unit) to CI: it finishes in ~15 minutes where a laptop takes an hour or more
• My PR passes all required CI/CD checks (e.g., lint, schema.d.ts sync check, etc.)
• My PR's scope is as isolated as possible; it only solves 1 specific problem
• I have received a Greptile Confidence Score of at least 4/5 before requesting a maintainer review (Greptile reviews automatically once the PR is opened; only comment @greptileai to re-request a review after pushing changes)

Screenshots / Proof of Fix

Both sides run the scanner the workflow pins (osv-scanner v2.3.8, sha256-verified) with the repo's osv-scanner.toml, the same command .github/workflows/osv-scan.yml runs:

osv-scanner scan source --config osv-scanner.toml -L uv.lock -L ui/litellm-dashboard/package-lock.json

Before (2ce4e3f)

1. The scan exits 1 with six findings:

Total 2 packages affected by 6 known vulnerabilities (0 Critical, 1 High, 4 Medium, 1 Low, 0 Unknown) from 1 ecosystem.
6 vulnerabilities can be fixed.
| OSV URL                             | CVSS | ECOSYSTEM | PACKAGE | VERSION | FIXED VERSION | SOURCE  |
| https://osv.dev/GHSA-23w6-3w8w-8484 | 4.8  | PyPI      | pypdf   | 6.15.0  | 6.16.1        | uv.lock |
| https://osv.dev/GHSA-763m-79hh-57f2 | 4.8  | PyPI      | pypdf   | 6.15.0  | 6.16.1        | uv.lock |
| https://osv.dev/GHSA-jp53-mhqp-8xcg | 6.9  | PyPI      | pypdf   | 6.15.0  | 6.16.0        | uv.lock |
| https://osv.dev/GHSA-8423-8fgw-73vq | 6.9  | PyPI      | tornado | 6.5.7   | 6.5.8         | uv.lock |
| https://osv.dev/GHSA-mpf4-983q-p7j4 | 7.5  | PyPI      | tornado | 6.5.7   | 6.5.8         | uv.lock |
| https://osv.dev/GHSA-wwv5-g3v4-889x | 2.3  | PyPI      | tornado | 6.5.7   | 6.5.8         | uv.lock |

2. The latest osv-scan run on the staging branch fails the same way, minus the tornado advisory published after it ran (run)

After (f7b37e4)

1. uv lock --upgrade-package pypdf with the raised floor: Updated pypdf v6.15.0 -> v6.16.2; the lock diff touches only pypdf, tornado (from the earlier commit) and the relative exclude-newer stamp
2. The same scan exits 0:

Scanned uv.lock file and found 453 packages
Scanned ui/litellm-dashboard/package-lock.json file and found 879 packages
GHSA-w8v5-vhqr-4h9v and 2 aliases have been filtered out because: diskcache has no fixed release published; remove this entry once one exists
GHSA-h7x2-h6g9-p789 and 1 alias have been filtered out because: mlflow has no fixed release published; remove this entry once one exists
Filtered 3 vulnerabilities from output
No issues found

3. This PR's own osv-scan check passes at this commit (run)

Type

🐛 Bug Fix

Caveats (if any)

Low

• litellm[proxy-runtime] now needs pypdf 6.16.1 or newer; an environment pinning pypdf below that stops resolving

Final Attestation

• The tests check the right things, including the edge cases, and regressions in the respective real-world customer use-cases are not possible after this PR

---

Note

Low Risk
Patch-level dependency floor bumps with no code changes; main impact is stricter resolution for installs pinning older tornado or pypdf.

Overview
Raises dependency floors so osv-scan stops flagging six newly published advisories on locked versions.

tornado is bumped via [tool.uv] constraint-dependencies from >=6.5.6 to >=6.5.8; the lock resolves 6.5.8 (was 6.5.7). pypdf in the proxy-runtime extra moves from >=6.12.0 to >=6.16.1; the lock resolves 6.16.2 (was 6.15.0). uv.lock is refreshed accordingly (including the exclude-newer stamp).

No application code changes—only version policy and lockfile updates to clear CI and downstream vulnerability scans.

^{Reviewed by Cursor Bugbot for commit f7b37e4. Bugbot is set up for automated code reviews on this repo. Configure here.}

[mateo-berri]

bugbot run

[greptile-apps]

Greptile Summary

This PR raises two dependency floors and refreshes the lockfile to consume patched releases.

• Raises the proxy-runtime pypdf requirement and resolves pypdf 6.16.2.
• Raises the global Tornado constraint and resolves Tornado 6.5.8.
• Refreshes the lockfile’s generated package metadata and cutoff timestamp.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains.

Important Files Changed

Filename	Overview
pyproject.toml	Raises the pypdf and Tornado minimum versions consistently with the stated dependency-hardening goal.
uv.lock	Reflects both updated constraints and resolves the intended patched package versions with matching artifacts.

_{Reviews (2): Last reviewed commit: "Merge remote-tracking branch 'origin/lit..." | Re-trigger Greptile}

[codspeed-hq]

Merging this PR will not alter performance

✅ 31 untouched benchmarks

---

_{Comparing litellm_bump_tornado_658 (f7b37e4) with litellm_internal_staging (4049a07)}

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 1eea8e2. Configure here.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[mateo-berri]

bugbot run

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 647f32c. Configure here.}

[lzhan011]

OSV now also flags pypdf 6.15.0 under three advisories; updating pypdf to 6.16.2 alongside tornado restores a clean scan

[mateo-berri]

bugbot run

[mateo-berri]

@greptileai

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit f7b37e4. Configure here.}