Generative AI has made the competition for attention fiercer than ever. Omneky fights to help you stand out. Attention is all you need.

San Francisco, CA
Joined July 2018
Pinned Tweet
Today we're launching TASTE BENCH, Omneky's eval suite for the creative quality of AI-generated ads. For image ads, the top model, GPT Image 2.5 Sunburst, averaged 7.38 out of 10 on quality, yet only 54.2% of its ads were ready to run as-is. Across eight models, ready-to-run rates ranged from 22.0% to 54.2%. Every model got the same 59 briefs across four brands and five languages, and a blind panel of AI judges from OpenAI, Anthropic, and Google checked each ad for exact copy, faithful logos and products, safe zones, and fabricated claims. These are first-attempt numbers with our own review step turned off. We built TASTE BENCH because once you can measure taste, you can hill climb it. It covers image and video models, and you can explore every image ad, brief, judge explanation, and score at omneky.com/tastebench.
1
2
1
8
20,800
Omneky today announced that its advertising harness earned the highest ad score in the TasteBench evaluation, scoring 8.8 and outperforming all six tested ChatGPT and Gemini configurations. The next-best configuration, ChatGPT using GPT-6 Astra at medium reasoning, scored 7.6. The results highlight the role of the harness in creative production. A harness is the software surrounding a model that supplies context and directs its work. Omneky's advertising workflow reads brand information, develops art direction, provides product and logo references, and selects an image model. TasteBench compares the ads produced by these complete workflows. "Creative is one of the most important levers for growth, and the quality of the finished ad matters to every marketer," said Hikari Senju, founder and CEO of Omneky. "We built Omneky's harness to bring brand knowledge and creative direction into every generation. TasteBench gives marketers a way to compare the quality of what these systems actually produce." app.dealroom.co/news/feed/om…
1
1
3
252
Omneky claims its harness beats ChatGPT and Gemini on TasteBench. <옴니키 "챗GPT·제미나이보다 광고 감각 우월", 누가 채점했나> 1. 옴니키가 자사 하네스로 TasteBench 광고 점수 1위라고 보도자료 발표 2. 모델이 아니라 틀의 승리, 채점 설계자가 곧 판매자 3. 감각을 점수로 만들면 점수용 광고가 나옴. 전환율 전엔 발표일 뿐
1
2
2
52
Omneky retweeted
Omneky's harness outperformed ChatGPT and Gemini on ad creative quality. Creative is one of the most important levers for growth, and the quality of the finished ad matters to every marketer. We built Omneky’s harness to bring brand knowledge and creative direction into every generation. TasteBench gives marketers a way to compare the quality of what these systems actually produce.
Omneky Harness Outperforms ChatGPT and Gemini on TasteBench Ad Score Omneky scored 8.8 in our latest TasteBench evaluation, ahead of every tested ChatGPT and Gemini configuration. The next-highest score was 7.6. Seven configurations. 52 ad requests. Four brands. 364 ads. Watch the side-by-side comparison and explore the full matrix to see how each harness turns a brief into an ad. See the results: omneky.com/tastebench
2
1
4
252
Omneky Harness Outperforms ChatGPT and Gemini on TasteBench Ad Score Omneky scored 8.8 in our latest TasteBench evaluation, ahead of every tested ChatGPT and Gemini configuration. The next-highest score was 7.6. Seven configurations. 52 ad requests. Four brands. 364 ads. Watch the side-by-side comparison and explore the full matrix to see how each harness turns a brief into an ad. See the results: omneky.com/tastebench
2
1
1
7
637
Your best ad already has the hook, the pacing and the edit. Omneky Clone keeps all of that and swaps what you want to change. Here we gave Clone a 15-second ad and one photo of a new outfit. Played side by side, the shots, cuts and timing match the original, and the new outfit shows up in every scene. Upload any video, then swap the avatar, replace the apparel or insert your product. Powered by Seedance 2.5. Try it at omneky.com #AIAdvertising #GenerativeAI #VideoAds #CreativeAutomation
3
1
6
424
Omneky was named Top AI-Powered Ad Campaign Optimization Platform by Silicon India! Check out our profile here: siliconindia.com/omneky-2026
1
3
279
Omneky now connects to Shopify. Your ads keep promoting products that are about to sell out, while overstocked items sit without fresh creative. Connect your store and the Omneky agent plans around what's actually in stock, using your live catalog, inventory and orders. What the agent can do: 🛍️ Answer "what's selling this week?" from live orders 🛍️ Flag low-stock products before ads overspend 🛍️ Shift budget toward fast-moving products 🛍️ Push overstocked items with fresh creative made from your synced catalog Inventory movement in. Better marketing strategy out. Connect in Omneky under Settings → Connections → Commerce. Omneky gets read-only access to your products, inventory and orders. omneky.com #Shopify #Ecommerce #PerformanceMarketing
1
1
2
289
We never selected the Omneky plugin. ChatGPT Work picked it anyway: it found and saved a new product, made an image ad, changed its colors on request, generated a video ad, drafted a TikTok launch for approval, and pulled a performance report. Try it today: chatgpt.com/plugins/plugin_a…
1
1
200
さて本日はOmnekyでバナーを20個作ります。 @j_omneky
2
2
17
2,117
Omneky en ChatGPT (24 sep 2026): Agentic Harness de ads y GTM.Growth Agent que lanza campañas multi-canal y consulta performance desde el chat.
2
2
2
166
Omneky published TASTE BENCH: a public scoreboard for whether AI image models actually produce ready-to-run ads. Best first-attempt ship rate among eight models: 54.2% (GPT Image 2.5 Sunburst). Floor: 22%. badsignal.ai/stories/omneky-…
2
2
68
Omneky retweeted
Replying to @BrendanFoody
We just released a public benchmark on creative quality xitter.cf/hsenju/status/21033527…
Introducing TASTE BENCH, Omneky's eval suite for the creative quality of AI-generated ads across image and video models. Creative is the most important lever for growth. A blind panel of AI judges scores every ad on eight quality dimensions, including brand fit and typography, and checks it for 11 failures such as gibberish text, invented claims, and logo errors. We also track cost, latency, refusals, and judge agreement. See every result at omneky.com/tastebench.
1
3
248
Omneky retweeted
Introducing TASTE BENCH, Omneky's eval suite for the creative quality of AI-generated ads across image and video models. Creative is the most important lever for growth. A blind panel of AI judges scores every ad on eight quality dimensions, including brand fit and typography, and checks it for 11 failures such as gibberish text, invented claims, and logo errors. We also track cost, latency, refusals, and judge agreement. See every result at omneky.com/tastebench.
Today we're launching TASTE BENCH, Omneky's eval suite for the creative quality of AI-generated ads. For image ads, the top model, GPT Image 2.5 Sunburst, averaged 7.38 out of 10 on quality, yet only 54.2% of its ads were ready to run as-is. Across eight models, ready-to-run rates ranged from 22.0% to 54.2%. Every model got the same 59 briefs across four brands and five languages, and a blind panel of AI judges from OpenAI, Anthropic, and Google checked each ad for exact copy, faithful logos and products, safe zones, and fabricated claims. These are first-attempt numbers with our own review step turned off. We built TASTE BENCH because once you can measure taste, you can hill climb it. It covers image and video models, and you can explore every image ad, brief, judge explanation, and score at omneky.com/tastebench.
8
110
1
132
294,888