Instagram Reels blurry or low resolution: probe before posting
Instagram lists blurry, low-resolution Reels among those it may show less. Check width, height and fps with Sume video inspect, then look at full-size stills.

Instagram says Reels that are blurry or low resolution are among those you may see less often. The page gives no pixel threshold, so a number would be invented. What you can do is measure your own file and look at it: Sume's video inspect returns width, height, fps, video_codec and size_bytes, and video frames returns full-size stills so you can judge sharpness yourself. Instagram's help page renders in the browser and could not be fetched as plain text, so the wording here comes from the search excerpt of that page on 2026-10-08. Open the page to confirm before you rely on it.
Measure first
Video inspect reads one clip you host on media.sume.com. With no frames field it also returns eight mid-bin stills at a default long edge of 768 pixels, which is enough for a quick look but not for judging sharpness. For that, call video frames on the same clip: it returns durable image artifacts at the source frame size unless you send max_edge (16 to 2160), and format: "png" keeps them lossless.
| Probe field | What to look for |
|---|---|
| width, height | A vertical clip under 1080x1920 has fewer pixels than the full-screen Reels frame |
| fps | A very low rate looks choppy rather than blurry |
| video_codec, pix_fmt | H.264 with yuv420p is the common delivery format |
| size_bytes, duration_seconds | Divide to get an average bitrate; a very low one means heavy compression |
| rotation | A non-zero value can mean a sideways clip that gets re-oriented |
The arithmetic that finds soft files
Average bitrate is size times eight over duration. A 15-second clip of 3 MB is 24,000,000 bits over 15 seconds, about 1.6 Mbps. A 1080x1920 clip at 1.6 Mbps is thin; the same pixels at 8 Mbps would be 15 MB. This is a heuristic, not an Instagram rule, since the page lists no bitrate. It tells you which of ten files to open and look at first.
Pulling stills
Request three frames, one early, one mid and one late, because a clip can start sharp and fall apart on fast motion. at takes 1 to 24 values in seconds, each at least zero and below the duration. A submit always returns 202, so poll the job and then read the frames.
Zoom the PNG to 100% and look at text edges and faces. If they are soft in the source, nothing downstream fixes that. Trimming in exact mode re-encodes and will not add detail, and the output conform to a larger size will not either.
{
"video_url": "https://media.sume.com/artifacts/artf_demo/reel.mp4",
"at": [0.5, 7, 14],
"format": "png"
}What to change instead
If the source is the problem, re-export from the original edit at the full vertical frame, or regenerate the clip at a higher resolution tier where your model offers one. Sume's docs for these tools do not describe a video upscaler, so this post makes no claim about one. If the clip is the right size but compressed, check the codec and consider an exact trim only when you need a different container or frame rate.
Sources
Related posts
More in Use cases
- Instagram says muted Reels are shown less: check has_audio
Instagram's help page lists muted Reels among those you may see less often. Probe has_audio with Sume video inspect, then add sound with a Timeline render.
- Instagram Reels with text over most of the frame: cap caption size
Instagram lists Reels mostly covered by text among those it may show less. Bound Sume captions with font_size_ratio, safe_width_ratio and max_words.
- Korean speech captions: korean-ad style with language ko, $0.20
Caption Korean speech on Sume with language ko and a Hangul style such as korean-ad or black-outline. Latin styles return 400, not tofu boxes. $0.20 a job.
- Language school promo, 15 s at 9:16: Seedance 2.5, Wan or Kling
A 15-second 9:16 class teaser is $1.88 on Wan 3.0 720p, $3.15 on Kling 3 with sound and $8.67 on Seedance 2.5 720p. What each buys on Sume.
Written by Sume