Polimake

Video bitrate: from H.264 (2003) to AV1 (2018), the real per-platform recommendations and why there is no universal number

Video bitrate explained with the depth it deserves: what it means technically, the evolution of codecs (H.264 since 2003, H.265/HEVC 2013, VP9 2013, AV1 2018), the real recommendations from YouTube, Netflix, and other platforms, and why the right question isn't 'what is the best bitrate' but 'what is the right one for this specific case.'

· Platform

The team behind Polimake. We explore the intersection of technology, creativity, and automation.

Published:
Video bitrate: from H.264 (2003) to AV1 (2018), the real per-platform recommendations and why there is no universal number

The bitrate of a video is the amount of data per second the file takes up, typically measured in kilobits per second (kbps) or megabits per second (Mbps). It is the most direct factor in the relationship between visual quality and file size: the higher the bitrate, the more detail preserved but the heavier the file; the lower the bitrate, the lighter the file but the more visible the compression (artifacts, blocking, smoothing of detail).

The question "what is the best bitrate?" has no universal answer because it depends on at least six variables: resolution, framerate, content (how much motion there is), the codec used, the target platform, and the use case. This guide covers each variable, the real recommendations from major platforms in 2026, and the decision principles.

The evolution of codecs: what defines what's possible

Before getting to a specific bitrate, you have to understand which codec is being used. A codec (compressor-decompressor) is the algorithm that compresses the raw video data. The same bitrate produces very different quality depending on the codec.

MPEG-2 (1995). The standard that dominated DVDs and digital TV in the '90s and 2000s. Today it's legacy except in traditional broadcast.

H.264 / AVC (2003). Developed by ITU-T and ISO/IEC in collaboration. It's probably the most widely used video codec in the world in 2026 — it remains the standard for Blu-ray, HD broadcast, the first YouTube standard, and most web platforms. A reasonable balance between efficiency and universal compatibility. Its license (via MPEG LA) has been a source of cost for manufacturers for decades.

H.265 / HEVC (2013). High Efficiency Video Coding. Successor to H.264 with roughly 50% better compression for the same visual quality. Designed especially for 4K and HDR. Its adoption has been slower than expected due to complex and costly licensing issues. It remains the standard for HEIC on iPhones since 2017.

VP9 (2013). Google's open source codec, an alternative to HEVC with no licensing cost. Used heavily on YouTube. Efficiency comparable to HEVC in many cases.

AV1 (2018). Open source codec developed by the Alliance for Open Media (AOMedia) — a consortium that includes Google, Mozilla, Microsoft, Amazon, Netflix, Apple, Cisco, Intel, IBM, and Samsung. Designed specifically to avoid patent costs. Roughly 30-50% better compression than H.265/VP9 in many cases. Growing adoption since 2020-2021 with hardware support in modern chips.

H.266 / VVC (2020). Versatile Video Coding. Successor to HEVC. Roughly 50% better compression than its predecessor. Adoption was very slow up to 2026 due to licensing issues and a lack of widespread hardware support.

The practical takeaway for 2026:

  • H.264 remains the most universal standard. If you don't know which codec to use, H.264 works everywhere.
  • H.265 for 4K and HDR when the target platform supports it.
  • VP9 and AV1 dominate on platforms that control their own pipeline (YouTube, Netflix).
  • AV1 is growing quickly as the codec of the future.

The real recommendations by platform

The major platforms publish specific recommendations. In 2026:

YouTube (official recommendations, standard H.264 codec):

  • 1080p 30fps: 8 Mbps
  • 1080p 60fps: 12 Mbps
  • 2K (1440p) 30fps: 16 Mbps
  • 2K 60fps: 24 Mbps
  • 4K (2160p) 30fps: 35-45 Mbps
  • 4K 60fps: 53-68 Mbps
  • HDR: roughly 25% more

YouTube re-encodes everything with its own codecs (VP9, AV1) after upload, so the uploaded file is just input. The recommendation for a high upload bitrate is to have high-quality material before re-compression.

Vimeo (more lenient with upload bitrate):

  • 1080p: 10-20 Mbps recommended
  • 4K: 30-60 Mbps recommended

Instagram / Reels:

  • 9:16 vertical 1080×1920: around 5-8 Mbps is usually enough
  • Instagram compresses aggressively, so uploading a higher bitrate than recommended doesn't improve the final quality

TikTok:

  • 9:16 vertical: 5-10 Mbps. The platform compresses significantly.

X (Twitter):

  • 1080p: up to 25 Mbps on upload, but re-compressed to roughly 5 Mbps.

LinkedIn:

  • 1080p: 5-10 Mbps. The platform compresses aggressively to save bandwidth.

Web hero / self-hosting on your own server:

  • 1080p: 4-8 Mbps is usually adequate for web (with muted autoplay)
  • 4K: 15-25 Mbps if the audience justifies the bandwidth

OTT streaming (Netflix, etc.) uses adaptive bitrate streaming (ABR) with several simultaneous bitrates. Netflix has published its encoding ladder optimized by title and genre — for a typical 1080p drama, the levels range from roughly 1.5 Mbps to 5 Mbps. The efficiency comes from adapting to the user's connection.

The six factors that affect the right bitrate

1. Resolution. More pixels = more data. A 4K video has 4 times more pixels than 1080p, requiring roughly 4 times more bitrate for the same quality per area (although perception changes with viewing distance).

2. Framerate. More frames per second = more data. 60fps vs 30fps typically requires ~50-100% more bitrate (not strictly 2x because there is redundancy between frames).

3. Content motion. A static interview (little motion between frames) compresses much better than a fast action scene (a lot of changing information). The same bitrate produces visually different quality depending on the content.

4. Codec. As covered above: AV1 can deliver quality similar to H.264 with a substantially lower bitrate.

5. Target platform. Platforms compress the material you give them. Uploading 50 Mbps to TikTok doesn't produce a better final result than uploading 8 Mbps; the algorithm compresses to the range the platform uses internally.

6. Use case. Is it a master file to preserve (high bitrate)? Is it a final version for a specific audience (optimized bitrate)? Is it a preview/draft (low bitrate)? The intent determines the appropriate range.

Constant Bitrate (CBR) vs. Variable Bitrate (VBR)

Another technical component that affects the result:

CBR (Constant Bitrate). The bitrate is stable throughout the entire video. Each second uses roughly the same amount of data. Useful when transmission requires a predictable bitrate (live streaming, broadcasts where bandwidth is fixed).

VBR (Variable Bitrate). The bitrate adapts to the content. Static scenes use less data; scenes with a lot of motion use more. For the same average bitrate, VBR produces better perceived quality than CBR, but the file is less predictable in size.

Single-pass VBR vs. two-pass VBR. The latter analyzes the entire video first to allocate bitrate optimally. Slower to encode but produces better quality for the same target size.

For most uses in 2026: two-pass VBR is the optimal option when there is processing time available. CBR only when a predictable bitrate is required (live streaming).

How to choose bitrate in practice

For a brand or creator producing video regularly:

For the master archive (the version you keep to reuse and re-export):

  • 1080p: 30-50 Mbps in H.264, or ProRes/DNxHD for professional production.
  • 4K: 80-150 Mbps in H.264, or ProRes/DNxHD.
  • Important: the master must survive re-exports. Keeping high quality justifies the large files.

For web deliverables:

  • 1080p web hero: 5-8 Mbps in H.264.
  • 1080p for social media: 5-10 Mbps depending on the platform.
  • 4K web: 15-25 Mbps.
  • For email / presentations: 1-3 Mbps of compressed 720p or 1080p.

For uploading to platforms (YouTube, Vimeo): follow the official recommendations. Uploading more bitrate than recommended rarely improves the final quality after re-compression.

How to check whether your bitrate is right

Three practical tests:

1. Play it back in real conditions. Not just in the editor. Upload it to the target platform and see how it looks there. What looks perfect in Premiere may have visible artifacts after Instagram's compression.

2. Inspect on representative screens. A small phone can hide artifacts that are visible on a 4K monitor. If your audience watches on large screens, test there.

3. Compare A/B with a higher bitrate. If exporting at 5 Mbps and at 10 Mbps produces visually indistinguishable videos, 5 Mbps is enough. If the difference is noticeable, go higher.

Common bitrate mistakes

Assuming more bitrate = better final quality. On platforms that re-encode (YouTube, Instagram, TikTok), the upload bitrate only matters up to a certain threshold. Beyond that it doesn't produce a better result.

Using the same bitrate for everything. Fast-moving 4K needs much more bitrate than a static 1080p interview. Optimize case by case.

Exporting thousand-megabit files unnecessarily. For web, that's wasted bandwidth and slow loading. Calibrate to the final use.

Not keeping a high-quality master. Compression is destructive. If your only file is 5 Mbps and you later need to reuse it for HD, you've already lost quality.

Confusing bitrate with resolution. They are different things. 4K at 5 Mbps probably looks worse than 1080p at 5 Mbps because the data is more diluted across more pixels.

Ignoring the codec. Specifying "10 Mbps" without a codec is ambiguous. AV1 at 5 Mbps can look better than H.264 at 10 Mbps.

Uploading in CBR when VBR would be better. For on-demand video (not live streaming), VBR produces a better quality-to-size ratio.

Bitrate and creative operations

For a brand producing video at scale, defining bitrate standards by channel and by use is operational discipline. Without standards, every export is decided ad-hoc, files vary inconsistently, masters get lost, and per-channel versions aren't optimized.

That coordination is the discipline of creative operations: content production defines export presets by channel, brand management ensures visual consistency across versions, and Media stores high-quality masters for re-export when needed.

At Polimake, that logic lives across three surfaces: Studio coordinates multi-channel production, Studio produces with established presets, and Media maintains the reusable masters.


If you manage audiovisual production and you've arrived here looking for an answer about bitrate, the most useful thing you can take away is the combination of three ideas: the right bitrate depends on six variables (resolution, fps, motion, codec, platform, use), platforms re-encode what you upload (uploading higher than recommended rarely produces a better final result), and keeping high-quality masters is essential because compression is irreversible.

To complement this, what format will my video be delivered in covers the container decision, audio or video covers production priorities, and how long should a corporate video be covers complementary decisions.

Quick references