Video Codecs Explained: H.264 vs H.265 vs AV1 vs ProRes (2026)

A codec is the technology that compresses video for storage and decompresses it for playback — the word is short for "coder-decoder." H.264 is a widely compatible delivery choice; H.265/HEVC and AV1 can reduce bitrate at comparable quality, with savings and support depending on the encoder, settings, and device. ProRes is an editing and mastering codec with larger files. Crucially, a codec is not the same as a container: MP4 and MOV are containers (the box); the codec is the compression method inside.

Codecs are where "why is my file 40GB," "why won't this play," and "why does editing this footage stutter" all come from — and almost nobody explains them clearly. This is the deep, plain-English guide: what a codec actually is, the codec-vs-container confusion that trips up everyone, how the major codecs compare, and exactly which to use for editing, uploading, and archiving. Quick versions are in the video codec and bitrate glossary entries.

What is a codec?

A codec (coder-decoder) is software or hardware that compresses video into a smaller file and decompresses it for playback. Raw, uncompressed video is enormous — a few seconds of uncompressed 4K can be gigabytes — so essentially every video you've ever watched is compressed by a codec.

The codec's job is a balancing act between three things:

• File size — smaller is better for storage, upload, and streaming.
• Quality — how good it looks at a given file size.
• Compute — how much processing power it takes to encode and decode.

A "better" codec squeezes more quality into a smaller file — but usually costs more computing power to encode and sometimes to play back. That single trade-off explains almost every difference between H.264, H.265, AV1, and the rest.

Codec vs container vs format (the confusion that trips up everyone)

This is the single most important distinction in the whole topic, and it's the source of endless confusion: MP4 is not a codec.

• A codec (H.264, H.265, AV1, ProRes...) is the compression method — how the picture data is encoded.
• A container (MP4, MOV, MKV, WebM...) is the file wrapper — a box that holds the compressed video, the audio, subtitles, chapters, and metadata together.

Think of it like a shipping box (the container) holding a specific product (the codec-compressed video). A single ".mp4" file could contain H.264, H.265, or AV1 video inside. That's why "MP4" tells you almost nothing about the actual compression — two MP4 files can be wildly different.

So when someone asks "what format should I export?" the real answer has two parts: which container and which codec inside it.

How video compression actually works

Understanding two ideas explains why some codecs are great for uploading and terrible for editing.

Lossy compression. Most delivery codecs are lossy: they permanently discard data the eye is unlikely to miss (subtle color detail, imperceptible high frequencies) to shrink the file. More compression = smaller file but more quality loss. Lossless compression preserves the data exactly; visually lossless means losses are not readily visible under particular viewing conditions, not that no data was discarded.

Intra-frame vs inter-frame (the big one). This is why editing some footage stutters:

• Intra-frame (all-I / all-Intra): every single frame is compressed as a complete, standalone image (like a series of JPEGs). Files are generally larger, but a frame can be decoded without reconstructing earlier pictures. ProRes and DNxHR work this way — which is exactly why they're smooth to edit and scrub.
• Inter-frame (Long-GOP): many frames use predictions from other frames plus correction data, with periodically independently coded pictures. H.264, H.265, and AV1 commonly use inter-frame prediction, although codec families can also support intra-only encoding. Dependencies can make random access more demanding, especially without suitable hardware decoding.

This helps explain the common roles of delivery codecs (often inter-frame, prioritizing smaller files) and editing/intermediate codecs (often intra-frame, prioritizing random access). These are workflow choices, not a rule that delivery footage cannot be edited. Compression keyframes are also distinct from the property markers in animation keyframing.

The major video codecs compared

Here's the honest 2026 comparison of the codecs you'll actually encounter:

H.264 (AVC): the universal default

H.264 (also called AVC) is a common default for broad compatibility. Supported profiles, bit depths, levels, and resolutions still vary, so not every H.264 file plays on every device. YouTube recommends H.264 in MP4 for uploads; that is a useful destination-specific starting point, not a universal guarantee.

H.265 (HEVC): efficient but complicated

H.265/HEVC can achieve comparable quality at a lower bitrate than H.264, which is useful for 4K. There is no fixed percentage saving: footage, encoder implementation, quality settings, and encode time all matter. Licensing and device/browser support also affect deployment. Check HEVC support in the destination and the actual playback environment rather than assuming every MP4 player accepts it.

AV1: royalty-free and winning streaming

AV1 is an open codec from the Alliance for Open Media, designed for royalty-free use. It can offer efficient compression, but the saving relative to another codec depends on the benchmark, encoder, and settings. Encoding speed, encoder quality, and hardware support still matter. MDN's WebCodecs guide distinguishes broad AV1 decoder support from more limited hardware encoder support. Test those separately; an AV1-capable player does not imply an AV1-capable export path. For browser support, hardware encoding, and upload advice, see AV1 for YouTube and the web in 2026.

ProRes: the editing and mastering codec

Apple ProRes is an intra-frame editing codec designed for high-quality post-production and efficient random access, rather than minimum file size. Its variants trade data rate against fidelity; "visually lossless" is not mathematically lossless. YouTube accepts ProRes and DNxHR uploads. Uploading a master can avoid an extra delivery encode, while H.264 may save transfer time and storage. Choose based on the destination and your available bandwidth.

Where bitrate fits in

Codec and bitrate work together. Bitrate is how much data is used per second of video (measured in Mbps). For a given codec, higher bitrate = better quality and bigger files; lower bitrate = smaller files and more compression artifacts (blockiness, banding).

Two encoding modes control how that data is spent across the video (the bitrate (CBR vs VBR) glossary entry has the short version):

• CBR (Constant Bitrate): the same data rate throughout. Predictable file size; good for live streaming.
• VBR (Variable Bitrate): more data for complex scenes (fast motion, detail), less for simple ones (a static talking-head). Better quality-per-size; standard for uploads.

For the full breakdown, including recommended settings by platform, see CBR vs VBR explained.

A more efficient codec (AV1, HEVC) achieves the same quality as H.264 at a lower bitrate — that's literally what "more efficient" means. This is also why the frame rate and resolution you choose affect the bitrate you need: more pixels and more frames per second need more data to stay clean.

Which codec should you use?

Match the codec to the job, not to "which is newest." Distinguish the platform's accepted upload formats from the codecs a viewer's device plays directly: platforms such as YouTube re-encode uploads.

The rule of thumb: use proxies or an intra-frame intermediate when native footage is difficult to edit, and upload a high-quality format the destination accepts. Native H.264/HEVC editing can work well with appropriate hardware. H.264 in MP4 is a convenient upload choice; ProRes, HEVC, or AV1 may suit a particular workflow, but check destination support before exporting.

AV1 for YouTube and the web in 2026

AV1 plays in current Chrome, Edge, and Firefox, and in Safari on Apple devices that have an AV1 hardware decoder. Encoding is the catch: it is slow without a recent hardware encoder. For YouTube and social uploads, H.264 or HEVC is usually the simpler export, because platforms such as YouTube re-encode uploads anyway.

Which browsers and devices can play AV1?

AV1 playback is broad on the web but not universal. According to caniuse, as of October 2026:

• Chrome (desktop and Android), Firefox, and current Edge versions support AV1 playback.
• Safari on macOS and iOS supports AV1 only on devices with a hardware AV1 decoder, such as iPhone 15 Pro and M3 MacBook Pro models. Older Apple devices need an H.264 or HEVC fallback.
• TVs, streaming boxes, and older phones vary by model and chip. Check the specific device rather than assuming.

Support can also depend on the profile, resolution, and bit depth. Web apps can test an exact configuration with WebCodecs, as described in why hardware acceleration matters.

Can your hardware encode AV1?

Hardware AV1 encoding is newer and less common than AV1 decoding. NVIDIA says its Ada Lovelace architecture (GeForce RTX 40-series) added AV1 encoding, while the earlier Ampere generation added AV1 decoding only. AMD's spec page for the Radeon RX 7900 XTX lists AV1 encode (both as of October 2026). Many laptops, phones, and older GPUs can play AV1 but cannot encode it in hardware.

Without a hardware encoder, an AV1 export falls back to a software encoder, which is usually slower than an H.264 export at similar settings. If an AV1 export takes far longer than expected, that is the first thing to check.

Encode time vs file size vs compatibility

These ratings are relative and typical, at comparable visual quality. They are not measurements: real results depend on the encoder, settings, hardware, and footage.

When should you export AV1 instead of H.264 or HEVC?

Start with one fact: platforms re-encode what you upload. YouTube's partner guidance says it always re-encodes videos to optimize playback quality. The codec your viewers receive is the platform's choice. Exporting AV1 does not guarantee viewers get AV1, and exporting H.264 does not limit them to H.264.

So for uploads, the goal is a clean, high-quality source that the platform accepts, delivered in a reasonable time:

• Export H.264 for the safe default. YouTube's recommended upload settings specify H.264 in MP4, and most platforms and editors handle it.
• Export HEVC when the destination accepts it and you want smaller high-resolution files, or your camera and editor already work in HEVC.
• Export AV1 when the destination explicitly accepts it, you have a hardware AV1 encoder (or time for a software encode), and a smaller upload matters. Check the platform's current supported-formats list first.
• For self-hosted web video, AV1 can reduce bandwidth, but offer an H.264 or HEVC fallback for browsers and devices that cannot decode it.

Whatever you choose, do not starve the upload of bitrate to save a few megabytes. The platform encodes again from your file, so artifacts in the upload carry into every version viewers see.

Why hardware acceleration matters

Codecs can be encoded and decoded in software (using the CPU) or with hardware acceleration (using dedicated chips in the GPU). Hardware acceleration is dramatically faster and more power-efficient — it's why your phone can play 4K HEVC without melting.

This matters for browser-based editing. WebCodecs exposes available encoders and decoders, which may use hardware or software. Support depends on the browser, OS, device, codec profile, and configuration; apps can check it with VideoDecoder.isConfigSupported and the encoder equivalent. WebGPU can support effects processing, but it does not make unsupported codecs available. For a deeper look at how browsers expose codecs (WebCodecs), see our explainer.

How to choose your export codec settings

When you export, here's the decision sequence.

Step 1: Decide delivery vs editing

For an upload, check the destination's accepted formats and recommended settings. For an intermediate, consider ProRes/DNxHR or proxies. A high-quality master may also be a valid upload; the trade-off is file size and transfer time.

Step 2: Pick the codec for your audience

For broad compatibility, consider H.264. For efficient 4K delivery or streaming, consider H.265 or AV1 only where the upload destination or direct playback environment supports them.

Step 3: Choose the container

MP4 for delivery (holds H.264/H.265/AV1). MOV for ProRes editing masters. WebM for VP9/AV1 web.

Step 4: Set the bitrate

Use VBR for uploads, and set a bitrate appropriate to your resolution and frame rate (platforms publish recommended bitrates). Higher for 4K and 60fps; lower for 1080p talking-heads.

Step 5: Use a platform preset when possible

The safest path is an export preset that already bundles the right codec, container, resolution, frame rate, and bitrate for each destination — so you don't hand-tune every field.

Common codec mistakes

• Confusing MP4 (container) with a codec. MP4 is the box; H.264/H.265/AV1 is what's inside.
• Ignoring native-editing performance. Test Long-GOP footage on your system; use proxies or intermediates if scrubbing is slow.
• Choosing upload format without checking requirements. ProRes is accepted by some platforms, including YouTube, but its transfer cost may outweigh the benefit.
• Using HEVC/AV1 where compatibility matters without a fallback — an old device may not decode it. H.264 is the safe default.
• Wrong bitrate. Too low = blocky artifacts; too high = huge files with no visible benefit.
• Ignoring hardware support. Software-encoding AV1 on a weak machine is painfully slow; use hardware where available.
• Assuming newer = always better. AV1 is more efficient but not always the right choice for a compatibility-first upload.

How AI and browser-based editors handle codecs

For most creators, codecs should be invisible — and increasingly they are. The goal of modern tools is that you never manually pick "H.264 High Profile, Level 4.2" again.

Some browser-based AI video editors use WebCodecs for supported source and export configurations. Others use software or server-side processing. Neither the word "AI" nor an MP4 extension guarantees compatibility. Export presets can simplify configuration, but their settings still need to match the destination's current guidance.

For Loopdesk (disclosure: Loopdesk is our product), use the feature specifications as the starting point for supported imports and MP4/H.264 delivery. This guide does not establish an exhaustive codec/profile/browser support matrix or a fallback for unsupported files. At the export-and-delivery stage, check a representative source and the exported file rather than assuming every codec is handled locally. Codecs are one stop on the full video editing glossary.

Frequently Asked Questions

What is a video codec?

A video codec (coder-decoder) is the technology that compresses video into a smaller file for storage and streaming, and decompresses it for playback. Examples include H.264, H.265/HEVC, AV1, and ProRes. Nearly every video you watch is compressed by a codec.

Is MP4 a codec?

No. MP4 is a container (a file wrapper) that holds compressed video, audio, and metadata together. The codec is the compression method inside the MP4 — usually H.264, but it could be H.265 or AV1. Two MP4 files can use completely different codecs.

What is the difference between H.264 and H.265?

H.265/HEVC can achieve comparable quality at a lower bitrate than H.264, but the saving depends on the footage, encoder, and settings. H.264 generally has broader compatibility; HEVC support and licensing need checking for the intended workflow.

Is AV1 better than H.265?

AV1 can offer efficient compression and was designed for royalty-free use, but it is not universally better. Compare quality, encode time, and support for your actual settings. Decoder availability does not guarantee hardware encoder availability.

What about AV1?

AV1 is an efficient, royalty-free codec. Current Chrome, Edge, and Firefox play it; Safari does on Apple devices with an AV1 hardware decoder. Encoding is slow without a hardware AV1 encoder. For uploads, H.264 or HEVC is usually simpler, since platforms such as YouTube re-encode uploads anyway.

What codec should I use for YouTube?

YouTube recommends H.264 in MP4 and also accepts formats including ProRes and HEVC. Upload a high-quality file in a supported format. A ProRes master can avoid an extra delivery encode; H.264 usually reduces upload size and transfer time.

Why is ProRes so big?

ProRes compresses each frame independently and is designed for post-production rather than minimum file size. Its variants trade data rate against fidelity. It can be used for editing, mastering, and uploads where accepted; visually lossless is not mathematically lossless.

What is the difference between a codec and bitrate?

The codec is the compression method; bitrate is how much data is used per second. For a given codec, higher bitrate means better quality and bigger files. A more efficient codec achieves the same quality at a lower bitrate.

Why does my footage stutter when I edit it?

Long-GOP decoding can be demanding, but storage speed, effects, resolution, and hardware support also matter. Check the bottleneck before transcoding. Proxies or intermediates may help; a browser editor is not automatically a cure.

---

Planning a browser-based edit? Check the Loopdesk import and export specifications, then open Loopdesk and verify a representative source and export before a larger job.