---
id: "CVE-2026-125259"
page: "https://infrastack.ai/live/CVE-2026-125259"
kind: "vulnerability"
aliases: ["CVE-2026-125259"]
title: "FFmpeg: integer overflow in ProRes encoder corrupts heap memory"
severity: "high"
cvss: 8.5
epss: null
exploited: false
status: "open"
category: "package"
products: [{"name": "ffmpeg", "fixed": [], "affected": "<= 9.0.2", "ecosystem": "vendor"}]
ecosystems: []
tags: ["ffmpeg", "heap-overflow", "package", "prores", "prores_ks", "video-transcoding"]
not_affected_if: ["no-prores-ks-encoding", "no-prores_ks-encoding", "trusted-input-only"]
published_at: "2026-11-09T20:18:12+00:00"
first_seen_at: "2026-11-09T20:48:13+00:00"
updated_at: "2026-11-09T20:50:55+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-125259", "first_seen_at": "2026-11-09T20:48:13+00:00"}, {"source": "analyst", "url": null, "first_seen_at": "2026-11-09T20:50:55+00:00"}]
exploit_sources: []
---

# FFmpeg: integer overflow in ProRes encoder corrupts heap memory

Link for people: https://infrastack.ai/live/CVE-2026-125259 (this .md is for agents)

FFmpeg miscalculates the buffer size for ProRes video encoding when frames are very large, because it uses 32-bit arithmetic. An attacker who supplies a high-resolution video transcoded with the prores_ks encoder can write past an undersized packet buffer and corrupt heap memory.

_Summary, check and fix written by a model from the sources; labelled as such._

## Affected

- vendor:ffmpeg affected <= 9.0.2 fixed no fixed version listed

## How to check

Run ffmpeg -version; releases through 9.0.2 are listed as affected. Only matters if you transcode untrusted video with the prores_ks encoder (check scripts/services for -c:v prores_ks or -c:v prores_ks via libavcodec). Check whether your build's proresenc_kostya_common.c has the 'Frame too large to encode' check.

## Fix

No fixed release named in sources. Fixed in FFmpeg git commit 439b452f (avcodec/proresenc_kostya: compute frame size upper bound in 64 bits); build from a source tree containing it or wait for a release newer than 9.0.2.

## Timeline

- 2026-11-09T20:48:13+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-125259
- 2026-11-09T20:50:55+00:00 analyst — Read commit (16+/9-: 64-bit bound, new EINVAL check on oversize frames, credit Joshua Rogers/AISLE) and VulnCheck advisory (CVSS 8.5, AV:L, UI:P, <=9.0.2). No fixed release stated; no exploitation or PoC stated; code execution not demonstrated, only heap corruption. web-cli out of Exa credits, used WebFetch; no social search done.

## Documents

- [advisory] Third-party advisory: www.vulncheck.com https://www.vulncheck.com/advisories/ffmpeg-through-9.0.2-heap-buffer-overflow-via-prores-prores-ks-encoder
- [fix] Patch: github.com https://github.com/FFmpeg/FFmpeg/commit/439b452f9fe916a9f7d8788ab6afee8fac846b0b

## From the sources

FFmpeg through 9.0.2 contains an integer overflow vulnerability in ff_prores_kostya_encode_init() that miscalculates frame_size_upper_bound using 32-bit arithmetic for large frame dimensions. Attackers can supply a high-resolution video transcoded with the prores_ks encoder, causing encode_frame() to write slices past an undersized packet buffer and corrupt heap memory.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_