{"id":983,"date":"2026-09-20T09:25:53","date_gmt":"2026-09-20T09:25:53","guid":{"rendered":"https:\/\/sebertech.com\/news\/?p=983"},"modified":"2026-09-20T09:25:54","modified_gmt":"2026-09-20T09:25:54","slug":"figure-helix-2-5-lets-humanoid-robots-do-chores-in-30-unseen-homes-with-56-success","status":"publish","type":"post","link":"https:\/\/sebertech.com\/news\/2026\/09\/20\/figure-helix-2-5-lets-humanoid-robots-do-chores-in-30-unseen-homes-with-56-success\/","title":{"rendered":"Figure Helix 2.5 Lets Humanoid Robots Do Chores in 30 Unseen Homes With 56% Success"},"content":{"rendered":"\n

Helix 2.5 marks progress toward humanoid robots that can work outside carefully prepared environments. But a 56% full-task success rate also shows that reliable home robots remain a work in progress.<\/strong><\/p>\n\n\n\n

\"Figure<\/figure>\n\n\n\n

Figure introduced Helix 2.5<\/strong> on September 17, 2026, saying its humanoids can perform household chores in homes they have never seen before without collecting new data or fine-tuning for those specific homes.<\/p>\n\n\n\n

The company tested the system in 30 Bay Area homes<\/strong>. Robots made beds, folded towels and tidied living rooms. Figure reported a 56% full-task success rate, compared with 9% for a policy trained without Index pretraining.<\/p>\n\n\n\n

What Figure Tested<\/h2>\n\n\n\n

Helix 2.5 was pretrained on Index<\/strong>, Figure’s large-scale dataset of human behaviour. The foundation model was then adapted to three chores: tidying living rooms, folding towels and making beds.<\/p>\n\n\n\n

The robots were evaluated in 30 homes that were not used for task training. Figure says no evaluation toy, towel or bedding appeared in the task-specification data. The robots also used furniture already present in each home.<\/p>\n\n\n\n

Success required completing the entire task. Tidying meant placing all 13 to 15 scattered toys into a basket. Towel folding required folding the towels and putting them in a basket. Bed making required completing the bedding arrangement under Figure’s predefined criteria. Partial completion did not count.<\/p>\n\n\n\n

What \u201cZero-Shot\u201d Means<\/h2>\n\n\n\n

Figure’s use of \u201czero-shot\u201d does not mean the robots had never been trained to make beds, fold towels or tidy rooms.<\/p>\n\n\n\n

The three behaviours were specified using fine-tuning data collected elsewhere. Zero-shot refers to the evaluation homes and manipulated objects<\/strong>. The robots received no additional adaptation after arriving in those homes.<\/p>\n\n\n\n

That makes the test a measure of whether previously learned physical skills can transfer to unfamiliar environments.<\/p>\n\n\n\n

How Well Did Helix 2.5 Perform?<\/h2>\n\n\n\n

Humanoids Daily, citing Figure’s evaluation chart, reported 237 successful trials out of 420<\/strong>, or about 56%.<\/p>\n\n\n\n

Bed making recorded 94 successes in 140 trials, or 67%. Towel folding reached 87 of 140, or 62%. Living-room tidying was the hardest task at 56 of 140, or 40%.<\/p>\n\n\n\n

Figure compared the Index-pretrained policy with one trained using the same task-specific data, architecture and evaluation setup but without Index pretraining. The company reported 56% success for the pretrained model and 9% for the baseline.<\/p>\n\n\n\n

Humanoids Daily noted that Figure’s chart shows 35 baseline successes in 420 trials, equal to about 8.3%. Figure’s written announcement states 9%, so the difference appears to be a small reporting or rounding discrepancy.<\/p>\n\n\n\n

Why the Result Matters<\/h2>\n\n\n\n

The main challenge is not teaching a robot to fold one towel or make one bed. It is getting the same learned behaviour to work across different rooms, furniture and objects without retraining for every location.<\/p>\n\n\n\n

Figure says Helix 2.5 used half as much task-specific adaptation data as a representative Helix 02 policy while generalising across 30 unfamiliar homes.<\/p>\n\n\n\n

The company also showed examples of robots recovering from mistakes by changing stance, repositioning or moving around furniture before continuing a task.<\/p>\n\n\n\n

Still, the reliability gap remains significant. About 44% of the 420 trials did not end in full task completion<\/strong> under Figure’s pass-or-fail criteria. The evaluation was also conducted by Figure rather than an independent testing organisation.<\/p>\n\n\n\n

No Consumer Launch Yet<\/h2>\n\n\n\n

Figure 03 was designed with home use in mind, but Figure has not announced a public consumer price, retail ordering programme or general household delivery date.<\/p>\n\n\n\n

That means Helix 2.5 is currently more relevant as a robotics research and development milestone than as a product consumers can buy.<\/p>\n\n\n\n

Figure Is Scaling Data and Compute<\/h2>\n\n\n\n

Figure is investing heavily in the infrastructure behind Helix. On September 3, it announced a partnership with Nscale involving an initial $3.5 billion compute commitment<\/strong> and plans for up to 100,000 GPUs on Nvidia’s Vera Rubin platform<\/strong>.<\/p>\n\n\n\n

Figure also said Index is generating about 35 minutes of new human-experience data every second<\/strong>.<\/p>\n\n\n\n

The next test will be whether more data and computing power can raise reliability, expand the range of household tasks and produce results that can be independently reproduced.<\/p>\n\n\n\n

Sources:<\/strong> Figure, Helix 2.5: Zero-Shot 30-Home Generalization<\/em>; Humanoids Daily, Figure\u2019s Helix 2.5 Takes on Chores in 30 Unseen Homes<\/em>; Figure, Figure and Nscale Sign Strategic Partnership<\/em><\/p>\n\n\n\n

Written by Liam Hisona<\/strong>
Published:<\/strong> September 18, 2026, 8:35 PM PHT<\/p>\n","protected":false},"excerpt":{"rendered":"

Helix 2.5 marks progress toward humanoid robots that can work outside carefully prepared environments. But a 56% full-task success rate also shows that reliable home … <\/p>\n","protected":false},"author":4,"featured_media":996,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[575,571,573,572,574],"class_list":["post-983","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-news","tag-figure","tag-figure-helix-2-5","tag-helix-2-5","tag-humanoid-robots","tag-humanoids"],"_links":{"self":[{"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/posts\/983","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/comments?post=983"}],"version-history":[{"count":2,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/posts\/983\/revisions"}],"predecessor-version":[{"id":997,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/posts\/983\/revisions\/997"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/media\/996"}],"wp:attachment":[{"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/media?parent=983"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/categories?post=983"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sebertech.com\/news\/wp-json\/wp\/v2\/tags?post=983"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}