Beamr (NASDAQ: BMR) links CABR video compression to stronger AI models

Rhea-AI Filing Summary

Beamr Imaging Ltd. reported research showing its patented Content-Adaptive Bitrate (CABR) video compression can improve AI training for machine vision. A state-of-the-art monocular depth estimation model fine-tuned on AV video compressed with CABR achieved a 35.2% file-size reduction relative to baseline compression.

The fine-tuned model showed a 30.7% reduction in depth estimation error on vulnerable road users, such as pedestrians and motorcyclists, and a 16.0% aggregate error reduction across all object classes. Prior Beamr benchmarks also showed up to 50% file-size reduction while preserving object detection accuracy at mean average precision of 0.96, and 41%–57% reduction for captioning workflows without measurable impact on outputs.

Key Figures

File-size reduction vs baseline: 35.2% reduction Depth error reduction on vulnerable users: 30.7% reduction Aggregate depth error reduction: 16.0% reduction +3 more

6 metrics

File-size reduction vs baseline 35.2% reduction CABR-compressed AV training data relative to baseline compression

Depth error reduction on vulnerable users 30.7% reduction Depth estimation error on pedestrians and motorcyclists

Aggregate depth error reduction 16.0% reduction Aggregate across all object classes in evaluated model

Prior max file-size reduction Up to 50% Content-adaptive compression in AV development pipeline benchmarks

Object detection accuracy 0.96 mAP Preserved mean average precision in prior AV benchmarks

Captioning workflow reduction 41%–57% reduction File-size reduction in captioning workflows with no measurable output impact

Key Terms

Content-Adaptive Bitrate (CABR), machine vision, monocular depth estimation, mean average precision, +1 more

5 terms

Content-Adaptive Bitrate (CABR) financial

"Beamr’s patented Content-Adaptive Bitrate (CABR) technology are more resilient"

Content-adaptive bitrate (CABR) is a video-delivery method that adjusts streaming quality in real time based on both the viewer’s network conditions and the complexity of what's on-screen, so simple scenes use less data while complex scenes get more detail. For investors, CABR matters because it can improve user experience and reduce bandwidth costs, which affects subscriber retention, platform scalability, and the economics of streaming services.

machine vision technical

"machine vision models fine-tuned on video compressed by Beamr’s patented"

Machine vision is technology that allows computers and machines to see, interpret, and analyze visual information from the world, much like how human eyes and brains work together. It enables automated systems to recognize objects, read signs, or inspect products without human help. For investors, machine vision is important because it drives advancements in automation and artificial intelligence, influencing industries and business efficiency.

monocular depth estimation technical

"a state-of-the-art monocular depth estimation model"

Monocular depth estimation is a computer vision technique that uses a single camera image to estimate how far away objects are, like judging distance from a photograph instead of using stereo eyes or sensors. For investors, it matters because the quality and cost of products that rely on cameras—such as autonomous vehicles, drones, robotics, augmented reality, and medical imaging—can be transformed by more accurate, cheaper depth sensing, affecting competitiveness, safety, and potential market adoption.

mean average precision technical

"while preserving object detection accuracy at mean average precision of 0.96"

Mean average precision is a single-number measure used to judge how well a ranking or recommendation system orders relevant items near the top; it averages precision scores at each point a correct item appears and then takes the mean across multiple queries. For investors, it matters when assessing tools that sort news, signals, or trade ideas — higher values mean the system more reliably surfaces the most important items first, like a well-organized playlist that puts your favorite songs at the top.

world foundation model pipelines technical

"captioning workflows in world foundation model pipelines showed 41%–57%"

Form 6-K: How Foreign Companies Report to the SEC →

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

Washington, D.C. 20549

Form 6-K

Report of Foreign Private Issuer

Pursuant to Rule 13a-16 or 15d-16

under the Securities Exchange Act of 1934

For the month of May 2026

Commission file number: 001-41523

BEAMR IMAGING LTD.

(Translation of registrant’s name into English)

10 HaManofim Street

Herzeliya, 4672561, Israel

(Address of principal executive offices)

Indicate by check mark whether the registrant files or will file annual reports under cover of Form 20-F or Form 40-F.

Form 20-F ☒           Form 40-F ☐

CONTENTS

Attached hereto and incorporated herein is the Registrant’s press release issued on May 6, 2026, titled “Beamr Research Validates Patented CABR Technology as an AI Training Asset”.

1

EXHIBIT INDEX

Exhibit No.
99.1		Press release titled: “Beamr Research Validates Patented CABR Technology as an AI Training Asset”.

2

SIGNATURES

Pursuant to the requirements of the Securities Exchange Act of 1934, the registrant has duly caused this report to be signed on its behalf by the undersigned, thereunto duly authorized.

	Beamr Imaging Ltd.
Date: May 6, 2026	By:	/s/ Sharon Carmel
	Name:	Sharon Carmel
	Title:	Chief Executive Officer

3

Exhibit 99.1

Beamr Research Validates Patented CABR Technology as an AI Training Asset

Training AI model on video data processed by Beamr’s content-adaptive technology made the model more resilient to compression, by lowering depth estimation error on safety-critical road users, including pedestrians and motorcyclists, by 30.7%

Herzliya, Israel, May 06, 2026 (GLOBE NEWSWIRE) -- Beamr Imaging Ltd. (NASDAQ: BMR), a leader in video optimization technology and solutions, released research demonstrating that machine vision models fine-tuned on video compressed by Beamr’s patented Content-Adaptive Bitrate (CABR) technology are more resilient than models trained on uncompressed data, while reducing the video data volumes that machine vision development depends on.

Machine vision teams handling petabyte-scale video data for autonomous vehicles (AV) and other video AI applications typically consider compression as a process for managing this scale. The findings reframe adaptive compression as an asset that strengthens AI model resilience, with the advantages of reducing storage and networking costs and infrastructure. This research extends Beamr’s ML-Safe benchmarks, validating a potential performance asset for AI models trained across machine vision applications.

The research evaluated Depth Anything V2, a state-of-the-art monocular depth estimation model. The model was fine-tuned on AV video data compressed with Beamr’s technology that delivered 35.2% file-size reduction relative to baseline compression. The fine-tuned model demonstrated 30.7% reduction in depth estimation error on vulnerable road users, including pedestrians and motorcyclists, and 16.0% aggregate reduction across all object classes. Full methodology and results are available in the blog post.

“This research shows that compressed video data can produce models that are more robust, not less,” said Dani Megrelishvili, Beamr CPO. “That points to a different role for compression in our customers’ pipelines, from a cost they tolerate to a tool they deploy.”

“Machine vision teams have faced a structural trade-off: compress video data to manage scale, or face the escalating costs and infrastructure challenges of running AI models without compression,” said Ronen Nissim, ML Lead at Beamr. “Our research suggests this trade-off is more flexible than the industry may have assumed. By using compressed footage as augmentation during fine-tuning, we produced a model that performed better on the validation set than the equivalent model trained on uncompressed data.”

Beamr’s ML-safe benchmarks have previously validated content-adaptive compression across the AV development pipeline. The benchmarks demonstrated up to 50% file size reduction while preserving object detection accuracy at mean average precision of 0.96, with high fidelity across detection, localization, and confidence consistency. Subsequent testing for captioning workflows in world foundation model pipelines showed 41%–57% file size reduction with no measurable impact on the pipeline outputs.

To run Beamr’s compression on your own data, visit beamr.com/autonomous

About Beamr

Beamr (Nasdaq: BMR) is a world leader in content-adaptive video compression, trusted by top media companies including Netflix and Paramount. Beamr’s perceptual optimization technology (CABR) is backed by 53 patents and a winner of Emmy® Award for Technology and Engineering. The innovative technology reduces video file sizes by up to 50% while preserving quality and enabling AI-powered enhancements.

Beamr powers efficient video workflows across high-growth markets, such as media and entertainment, user-generated content, machine learning, and autonomous vehicles. Its flexible deployment options include on-premises, private or public cloud, with convenient availability for Amazon Web Services (AWS) and Oracle Cloud Infrastructure (OCI) customers.

For more details, please visit www.beamr.com or the investors’ website www.investors.beamr.com

Forward-Looking Statements

This press release contains “forward-looking statements” that are subject to substantial risks and uncertainties. Forward-looking statements in this communication may include, among other things, statements about Beamr’s strategic and business plans, technology, relationships, objectives and expectations for its business, the impact of trends on and interest in its business, intellectual property or product and its future results, operations and financial performance and condition. All statements, other than statements of historical fact, contained in this press release are forward-looking statements. Forward-looking statements contained in this press release may be identified by the use of words such as “anticipate,” “believe,” “contemplate,” “could,” “estimate,” “expect,” “intend,” “seek,” “may,” “might,” “plan,” “potential,” “predict,” “project,” “target,” “aim,” “should,” “will” “would,” or the negative of these words or other similar expressions, although not all forward-looking statements contain these words. Forward-looking statements are based on the Company’s current expectations and are subject to inherent uncertainties, risks and assumptions that are difficult to predict. Further, certain forward-looking statements are based on assumptions as to future events that may not prove to be accurate. For a more detailed description of the risks and uncertainties affecting the Company, reference is made to the Company’s reports filed from time to time with the Securities and Exchange Commission (“SEC”), including, but not limited to, the risks detailed in the Company’s annual report filed with the SEC on February 26, 2026 and in subsequent filings with the SEC. Forward-looking statements contained in this announcement are made as of the date hereof and the Company undertakes no duty to update such information except as required under applicable law.

Investor Contact:
investorrelations@beamr.com

FAQ

What did Beamr (BMR) announce in its May 2026 Form 6-K?

Beamr announced research validating its CABR video compression as an AI training asset. A depth estimation model trained on CABR-compressed AV video achieved 35.2% file-size reduction and materially lower depth estimation errors, especially on vulnerable road users like pedestrians and motorcyclists.

How does Beamr’s CABR technology affect AI model performance for BMR?

Beamr reports that CABR-compressed training data improved AI resilience. A fine-tuned monocular depth estimation model showed a 30.7% reduction in depth estimation error on vulnerable road users and a 16.0% aggregate error reduction across all object classes versus a model trained on uncompressed data.

What file-size reductions does Beamr (BMR) claim for its CABR compression?

In the new research, CABR delivered a 35.2% file-size reduction relative to baseline compression. Earlier ML-safe benchmarks showed up to 50% file-size reduction while preserving object detection accuracy, and 41%–57% reduction in captioning workflows with no measurable impact on pipeline outputs.

Which AI model was evaluated in Beamr’s (BMR) CABR research?

Beamr evaluated Depth Anything V2, a state-of-the-art monocular depth estimation model. When fine-tuned on autonomous vehicle video compressed using CABR, the model achieved lower depth estimation errors and smaller video file sizes compared with training on uncompressed data, according to the company.

How is Beamr’s CABR technology positioned for autonomous vehicles and AI?

Beamr positions CABR as a tool for autonomous vehicle and video AI teams handling petabyte-scale data. The company states CABR can reduce storage and networking needs while maintaining or improving AI model performance, including resilience on safety-critical objects like pedestrians and motorcyclists.

What prior ML-safe benchmark results has Beamr (BMR) disclosed?

Beamr’s earlier ML-safe benchmarks showed content-adaptive compression delivering up to 50% file-size reduction while preserving object detection accuracy at mean average precision of 0.96, and 41%–57% size reduction in captioning workflows without measurable impact on world foundation model pipeline outputs.

Filing Exhibits & Attachments

1 document

Press Releases

• EX-99.1 PRESS RELEASE TITLED: "BEAMR RESEARCH VALIDATES PATENTED CABR TECHNOLOGY AS AN A 13.8 KB