What Is the Maximum Resolution librav1e Can Encode?

This article explores the theoretical maximum resolution that the librav1e AV1 video encoder can support. By examining the underlying AV1 bitstream specifications and the design of the Rust-based rav1e encoder, we identify the exact pixel boundaries, profile constraints, and practical limitations of this modern encoding technology.

The Theoretical Bitstream Limit

Because librav1e is the C-compatible library interface for the rav1e encoder, its limits are fundamentally bound by the AV1 video coding format specifications. The AV1 bitstream syntax defines the maximum frame width and frame height using 16-bit unsigned integers. Consequently, the absolute theoretical maximum resolution that librav1e can encode is 65,536 x 65,536 pixels (often referred to as 64K resolution) per frame. This equates to an image size of approximately 4.29 gigapixels.

AV1 Levels and Profile Constraints

While the bitstream syntax allows up to 64K resolution, the AV1 standard defines specific “Levels” to ensure hardware compatibility. Most hardware decoders and software players do not support the maximum theoretical limit of the bitstream.

For instance, the highest standardized tier in the AV1 specification is Level 6.3, which targets an 8K resolution of 7680 x 4320 pixels at 120 frames per second with a maximum bitrate of 800 Mbps. While librav1e can theoretically generate compliant streams up to the absolute 65,536 x 65,536 limit, any video encoded beyond the established level limits will fail to play on standard AV1-compliant hardware decoders.

Practical Hardware Limitations

In real-world scenarios, encoding a video at or near the 64K theoretical limit using librav1e is constrained by physical hardware capacity rather than software design.

In summary, while librav1e can theoretically encode video resolutions up to 65,536 x 65,536 pixels due to AV1 bitstream design, practical constraints limit standard deployment to 4K and 8K resolutions.