正在加载视频...

视频加载失败

返回首页

EVEN ARCHITECTURE STUDIO LANDING PAGES ARE GETTING SOLVED WITH A PROMPT → K3: careful typography, clear visual hierarchy → Opus: more generic layout, less brand personality K3: $0.20 in API. Opus 4.8: ~$0.40 design is getting solved by a prompt too

chesny's profile picture

Chesny

29,585 subscribers

91,196 次观看 • 2 个月前 •via X (Twitter)

科学技术
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

KIMI K3 VS OPUS 4.8 SIMULATING A 3D PARTICLE SYSTEM same prompt, same target: thousands of particles reacting to gravity and to each other in real time → K3: more organic distribution, smoother motion, particles actually cluster and drift like something physical → Opus: more rigid pattern, less natural, movement reads more like a grid than a simulation K3: $0.65 in API. Opus 4.8: ~$1.30 this is simulated physics, not just aesthetics, and it's the kind of detail that separates a demo from something you'd actually ship
0:14

KIMI K3 VS OPUS 4.8 SIMULATING A 3D PARTICLE SYSTEM same prompt, same target: thousands of particles reacting to gravity and to each other in real time → K3: more organic distribution, smoother motion, particles actually cluster and drift like something physical → Opus: more rigid pattern, less natural, movement reads more like a grid than a simulation K3: $0.65 in API. Opus 4.8: ~$1.30 this is simulated physics, not just aesthetics, and it's the kind of detail that separates a demo from something you'd actually ship

Jade

32,766 次观看 • 2 个月前

I tested Kimi K3 vs Claude Opus 4.8 Same prompt, an armory bay with lighting, props, and detail. Top is Kimi K3, bottom is Opus 4.8. It's not even close. Kimi K3 built a full scene with textures, proper lighting, ammo crates, weapon racks, working detail everywhere. Opus 4.8 gave me a near empty room with a couple of floating tables. No doubt it beats Opus 4.8. Kimi K3 is Fable 5 level, and it's clearly better than GPT-5.6 Sol at 3D and games. An open weight model just matched the best closed models on the market. Let that sink in.
0:13

I tested Kimi K3 vs Claude Opus 4.8 Same prompt, an armory bay with lighting, props, and detail. Top is Kimi K3, bottom is Opus 4.8. It's not even close. Kimi K3 built a full scene with textures, proper lighting, ammo crates, weapon racks, working detail everywhere. Opus 4.8 gave me a near empty room with a couple of floating tables. No doubt it beats Opus 4.8. Kimi K3 is Fable 5 level, and it's clearly better than GPT-5.6 Sol at 3D and games. An open weight model just matched the best closed models on the market. Let that sink in.

Bhavy☄️

479,495 次观看 • 2 个月前

KIMI K3 JUST RECREATED A FUNCTIONAL GAME BOY ADVANCE IN 3D → interactive 3D model, every button and screen modeled → a game loaded and playable right inside it → camera you can rotate and zoom around the device → one single prompt, zero external assets $3 in API with K3. the same result would run about $9 with Fable 5 and $6 with Opus 4.8 this isn't generating a static image of hardware anymore, it's generating a working simulation of it, from the shell to the screen to the game running inside
0:15

KIMI K3 JUST RECREATED A FUNCTIONAL GAME BOY ADVANCE IN 3D → interactive 3D model, every button and screen modeled → a game loaded and playable right inside it → camera you can rotate and zoom around the device → one single prompt, zero external assets $3 in API with K3. the same result would run about $9 with Fable 5 and $6 with Opus 4.8 this isn't generating a static image of hardware anymore, it's generating a working simulation of it, from the shell to the screen to the game running inside

alex

38,248 次观看 • 2 个月前

I'M PROBABLY ONE OF THE PEOPLE CONTRIBUTING TO THE GPUS DYING 😭 Built the same Crossy Road clone with both Kimi K3 and Opus 4.8 using their respective API keys inside Cursor, so neither model had a framework advantage despite the 2.7× difference in cost. Both produced fully playable games with solid physics, but Kimi's UI felt slightly more polished Kimi cost $7.11, Opus cost $19. Pretty impressive considering the price gap
0:15

I'M PROBABLY ONE OF THE PEOPLE CONTRIBUTING TO THE GPUS DYING 😭 Built the same Crossy Road clone with both Kimi K3 and Opus 4.8 using their respective API keys inside Cursor, so neither model had a framework advantage despite the 2.7× difference in cost. Both produced fully playable games with solid physics, but Kimi's UI felt slightly more polished Kimi cost $7.11, Opus cost $19. Pretty impressive considering the price gap

aditya

47,488 次观看 • 2 个月前

Blender MCP w/ Codex 5.5 or OPUS 4.8 is seriously pretty cracked... This entire block out was one-shot with a single prompt in about 15 minutes. Only problem I am having now is getting Seedance to use it properly 😅 Please feel free to use it and post your results below 👇
0:15

Blender MCP w/ Codex 5.5 or OPUS 4.8 is seriously pretty cracked... This entire block out was one-shot with a single prompt in about 15 minutes. Only problem I am having now is getting Seedance to use it properly 😅 Please feel free to use it and post your results below 👇

A.I.Warper

24,894 次观看 • 3 个月前

LongCat performed Opus 4.8 and GPT 5.5 level on real physics tasks for $0! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics Prompts: - A cannon demolishing a brick wall - A bowling ball knocking down the pins - A tornado that sucks in random objects Outputs: LongCat: 18,015 tokens, $0.00 Opus 4.8: 18,872 tokens, $0.48 GPT 5.5: 32,588 tokens, $0.98 GLM 5.2: 31,062 tokens, $0.09 On the physics LongCat came out ahead of Opus 4.8 and GLM 5.2 - cleaner collisions, nothing clipping or falling through. On detail and rendering it matched GPT 5.5, the best looking of the four. Getting this quality for free is wild!
0:16

LongCat performed Opus 4.8 and GPT 5.5 level on real physics tasks for $0! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics Prompts: - A cannon demolishing a brick wall - A bowling ball knocking down the pins - A tornado that sucks in random objects Outputs: LongCat: 18,015 tokens, $0.00 Opus 4.8: 18,872 tokens, $0.48 GPT 5.5: 32,588 tokens, $0.98 GLM 5.2: 31,062 tokens, $0.09 On the physics LongCat came out ahead of Opus 4.8 and GLM 5.2 - cleaner collisions, nothing clipping or falling through. On detail and rendering it matched GPT 5.5, the best looking of the four. Getting this quality for free is wild!

atomic.chat

105,523 次观看 • 3 个月前

🤯KIMI K3 ABSOLUTELY MOGS! BEATING Opus 4.8, GPT 5.5, and even Fable 5 in multiple benchmarks. They scored 1688 on GDPval-AA v2 🔥 This is a completely different breed of open-source models Kimi creates better games and front-end designs than Fable 5, but it's 8x cheaper! The results coming out are truly impressive! I will be testing this out further and posting multiple tests today. Stay tuned!
0:24

🤯KIMI K3 ABSOLUTELY MOGS! BEATING Opus 4.8, GPT 5.5, and even Fable 5 in multiple benchmarks. They scored 1688 on GDPval-AA v2 🔥 This is a completely different breed of open-source models Kimi creates better games and front-end designs than Fable 5, but it's 8x cheaper! The results coming out are truly impressive! I will be testing this out further and posting multiple tests today. Stay tuned!

Mark Santos

128,202 次观看 • 2 个月前

Okay, motion designers are 100% cooked. Saw on x that Opus 5.5 could do advanced animations and I was skeptical at first. So I gave it a one sentence prompt, and it one shot this in 15 minutes. Will add the prompt below, but this is genuinely mind blowing. This level of work would have costed $8,000 - $10,000 with a motion design team and 3-4 week turnaround with multiple revisions. Claude did it in one go with this prompt: "make a dynamic 15-second motion graphics video that shows what an incredible motion designer you are, like it's your showreel for a résumé. go all out." The only thing remaining for humans is creativity and critical thinking.
0:15

Okay, motion designers are 100% cooked. Saw on x that Opus 5.5 could do advanced animations and I was skeptical at first. So I gave it a one sentence prompt, and it one shot this in 15 minutes. Will add the prompt below, but this is genuinely mind blowing. This level of work would have costed $8,000 - $10,000 with a motion design team and 3-4 week turnaround with multiple revisions. Claude did it in one go with this prompt: "make a dynamic 15-second motion graphics video that shows what an incredible motion designer you are, like it's your showreel for a résumé. go all out." The only thing remaining for humans is creativity and critical thinking.

Leon Abboud

109,218 次观看 • 4 天前

Claude DESTROYS ChatGPT for building VSL funnels. I put together the Claude VSL Prompt Vault (below) I've booked 500+ calls for clients with VSL funnels. $1M generated. These prompts are how I build every single one. The same prompts behind every VSL funnel I've built since 2023. Scripts, landing pages, offer stacks, everything. This is the closest thing to hiring a funnel strategist for free. What's inside the vault: → Custom Claude System Prompt for VSL Scripts → The 5,280-Word VSL Conversion Framework → AI Audience Research Prompt (2-Part Deep Dive) → AI Competitor Funnel Analysis Prompt → AI Hook Generator (15+ Variations by Awareness Level) → AI Proof Stacking Prompt → AI Offer Architecture Builder (2-Part Sequence) → AI Landing Page Copy Prompt → AI VSL Script Audit + Rewrite Prompt Want access to this vault? → Comment "CLAUDE" → Follow and I'll DM the vault!
0:15

Claude DESTROYS ChatGPT for building VSL funnels. I put together the Claude VSL Prompt Vault (below) I've booked 500+ calls for clients with VSL funnels. $1M generated. These prompts are how I build every single one. The same prompts behind every VSL funnel I've built since 2023. Scripts, landing pages, offer stacks, everything. This is the closest thing to hiring a funnel strategist for free. What's inside the vault: → Custom Claude System Prompt for VSL Scripts → The 5,280-Word VSL Conversion Framework → AI Audience Research Prompt (2-Part Deep Dive) → AI Competitor Funnel Analysis Prompt → AI Hook Generator (15+ Variations by Awareness Level) → AI Proof Stacking Prompt → AI Offer Architecture Builder (2-Part Sequence) → AI Landing Page Copy Prompt → AI VSL Script Audit + Rewrite Prompt Want access to this vault? → Comment "CLAUDE" → Follow and I'll DM the vault!

Lucas H

13,804 次观看 • 6 个月前

Opus 4.6 vs GPT 5.4 (High) (1/9) prompt: Build a single-file HTML/CSS/JS (no libs) demo that uses SVG to simulate a plant growing: stem extends, leaves sprout + unfurl with springy/windy “physics”, then seamlessly loops forever. For the initial impressions I'm really impressed by GPT, for speed they both felt about the same, but gpt was still half cheaper than opus. I also much prefer the design and animation that gpt produced, physics on the leaves are super cool and it also loops pretty nicely whilst opus just fades out the plant. Still got a bunch of tests to run but this is really exciting.
0:10

Opus 4.6 vs GPT 5.4 (High) (1/9) prompt: Build a single-file HTML/CSS/JS (no libs) demo that uses SVG to simulate a plant growing: stem extends, leaves sprout + unfurl with springy/windy “physics”, then seamlessly loops forever. For the initial impressions I'm really impressed by GPT, for speed they both felt about the same, but gpt was still half cheaper than opus. I also much prefer the design and animation that gpt produced, physics on the leaves are super cool and it also loops pretty nicely whilst opus just fades out the plant. Still got a bunch of tests to run but this is really exciting.

Dev Ed

660,154 次观看 • 6 个月前

Been using Qwen 3.8 27B (Q4) locally on 64GB of VRAM. Here is the verdict: SLOW 18 tps with ZERO system prompt to process and that degrades significantly with a harness system prompt and as the context window grows. RIP if you have to compact. I had it implement this PRD and it's been running for 6 hours. By comparison Grok 4.6 and Kimi K3 hosted finished in about ~30 minutes. High hopes, but these 27B variants are too dense. This is not a consumer grade local model - and I consider consumer grade to be anything up to $5000.
0:17

Been using Qwen 3.8 27B (Q4) locally on 64GB of VRAM. Here is the verdict: SLOW 18 tps with ZERO system prompt to process and that degrades significantly with a harness system prompt and as the context window grows. RIP if you have to compact. I had it implement this PRD and it's been running for 6 hours. By comparison Grok 4.6 and Kimi K3 hosted finished in about ~30 minutes. High hopes, but these 27B variants are too dense. This is not a consumer grade local model - and I consider consumer grade to be anything up to $5000.

Burke Holland

122,428 次观看 • 1 个月前

Hailuo MiniMax (official) MiniMax Design (H3) H3 A prototype motion-graphics landing page for ice cream, built to test just how far H3 can push UI animation, typography, transitions, hover interactions, and product motion. I tried creating similar effects manually in Adobe After Effects but this version is 100% generated with H3. The model handles these complex motion-design elements incredibly well, and the results are STUNNING - Animated Shape Layers - Kinetic Typography - Cursor Hover Update - Bounce Effects - Wiggle Effects - Live Background - Micro Text Renders - Ease in Ease out Keyframes Prompt Below
0:15

Hailuo MiniMax (official) MiniMax Design (H3) H3 A prototype motion-graphics landing page for ice cream, built to test just how far H3 can push UI animation, typography, transitions, hover interactions, and product motion. I tried creating similar effects manually in Adobe After Effects but this version is 100% generated with H3. The model handles these complex motion-design elements incredibly well, and the results are STUNNING - Animated Shape Layers - Kinetic Typography - Cursor Hover Update - Bounce Effects - Wiggle Effects - Live Background - Micro Text Renders - Ease in Ease out Keyframes Prompt Below

Shams

35,225 次观看 • 1 个月前

Opus 5.5 just filed a pull request that changes the streaming behavior in T3 code. It's so cool that I can send off a prompt with a vague idea of what I want, and the result is a ready-to-merge pull request with a video demo of the changes directly in the PR. The PRs I make with AI are significantly better than the ones I used to make by hand. The future is awesome.
0:10

Opus 5.5 just filed a pull request that changes the streaming behavior in T3 code. It's so cool that I can send off a prompt with a vague idea of what I want, and the result is a ready-to-merge pull request with a video demo of the changes directly in the PR. The PRs I make with AI are significantly better than the ones I used to make by hand. The future is awesome.

Theo - t3.gg

92,784 次观看 • 5 天前

Codex cooked for 12 hours 19 minutes last night using a /goal 🎉 It added: - Stopwatch (count up) - Clock (with seconds) - Pomodoro - Tabata - Autocomplete Everything worked except the UI interactions around auto-complete (Codex GPT-5.5 is still not great at UI, but even Claude Opus 4.8 Ultracode struggled here). Overall, I'm really happy with the results. These are features that I've wanted for a while, and getting them close to release is awesome.
0:17

Codex cooked for 12 hours 19 minutes last night using a /goal 🎉 It added: - Stopwatch (count up) - Clock (with seconds) - Pomodoro - Tabata - Autocomplete Everything worked except the UI interactions around auto-complete (Codex GPT-5.5 is still not great at UI, but even Claude Opus 4.8 Ultracode struggled here). Overall, I'm really happy with the results. These are features that I've wanted for a while, and getting them close to release is awesome.

Paul Solt

48,887 次观看 • 3 个月前

I just built a Claude prompt library that runs your entire DTC marketing operation 🤯 100+ prompts organized by function: competitor research, creative briefs, ad copy, hooks, landing pages, performance analysis, customer review mining, and more. Perfect for DTC brands and agencies who are still prompting Claude from scratch every time they open a new chat, rewriting the same context, and getting generic output that sounds like every other AI-generated ad. This prompt library eliminates the entire loop: → Competitor Research: scrape and analyze competitor ads, extract winning hooks, map creative strategies, build competitive battlecards → Creative Briefs: generate data-backed briefs from ad performance, write iteration briefs, new concept briefs, test plans → Ad Copy & Hooks: 20 hooks across 10 frameworks, full ad copy variations, persona-specific angles, fatigue-busting rewrites → Landing Pages: audit any landing page against DR best practices, clone high-converting advertorial structures, write product page copy → Performance Analysis: audit Google Ads accounts, find wasted spend, build visual dashboards, weekly narrative reports → Customer Intelligence: mine reviews for ad copy language, extract objections, find unexpected use cases, build persona cards from real data → SEO & Content: find keyword gaps, write content in your brand voice, optimize product listings for AI shopping (ChatGPT, Gemini) → Email & SMS: launch sequences, weekly newsletters, abandoned cart flows, post-purchase nurture No more blank-page prompting. No more re-explaining your brand every session. No more generic AI output that sounds like a template. What you get: →100+ copy-paste prompts organized by the 8 functions DTC teams actually run →Every prompt pre-loaded with the context structure Claude needs to give you real output →Prompts that reference your brand voice, your ICPs, and your real data — not generic placeholders →A living library you can customize once and reuse across every campaign I put together the full prompt library as a single downloadable playbook: organized by section, ready to copy-paste into Claude today. Want it for free? > Like this post >Comment "PROMPTS" And I'll send it over (must be following so I can DM)
0:14

I just built a Claude prompt library that runs your entire DTC marketing operation 🤯 100+ prompts organized by function: competitor research, creative briefs, ad copy, hooks, landing pages, performance analysis, customer review mining, and more. Perfect for DTC brands and agencies who are still prompting Claude from scratch every time they open a new chat, rewriting the same context, and getting generic output that sounds like every other AI-generated ad. This prompt library eliminates the entire loop: → Competitor Research: scrape and analyze competitor ads, extract winning hooks, map creative strategies, build competitive battlecards → Creative Briefs: generate data-backed briefs from ad performance, write iteration briefs, new concept briefs, test plans → Ad Copy & Hooks: 20 hooks across 10 frameworks, full ad copy variations, persona-specific angles, fatigue-busting rewrites → Landing Pages: audit any landing page against DR best practices, clone high-converting advertorial structures, write product page copy → Performance Analysis: audit Google Ads accounts, find wasted spend, build visual dashboards, weekly narrative reports → Customer Intelligence: mine reviews for ad copy language, extract objections, find unexpected use cases, build persona cards from real data → SEO & Content: find keyword gaps, write content in your brand voice, optimize product listings for AI shopping (ChatGPT, Gemini) → Email & SMS: launch sequences, weekly newsletters, abandoned cart flows, post-purchase nurture No more blank-page prompting. No more re-explaining your brand every session. No more generic AI output that sounds like a template. What you get: →100+ copy-paste prompts organized by the 8 functions DTC teams actually run →Every prompt pre-loaded with the context structure Claude needs to give you real output →Prompts that reference your brand voice, your ICPs, and your real data — not generic placeholders →A living library you can customize once and reuse across every campaign I put together the full prompt library as a single downloadable playbook: organized by section, ready to copy-paste into Claude today. Want it for free? > Like this post >Comment "PROMPTS" And I'll send it over (must be following so I can DM)

Mike Futia

34,929 次观看 • 5 个月前

FABLE 5 JUST RECREATED NEW YORK CITY IN BLENDER. I connected Blender to Fable 5 and in like 20 minutes, it remade the entire cityscape of NYC It started by getting data of the buildings from public sources, THEN began building it (imo this is a smarter move than what Opus 4.8 would've done) So theoretically, this entire build is up to scale too Fable 5, welcome back.
0:18

FABLE 5 JUST RECREATED NEW YORK CITY IN BLENDER. I connected Blender to Fable 5 and in like 20 minutes, it remade the entire cityscape of NYC It started by getting data of the buildings from public sources, THEN began building it (imo this is a smarter move than what Opus 4.8 would've done) So theoretically, this entire build is up to scale too Fable 5, welcome back.

ashen

224,308 次观看 • 3 个月前

HERMES AGENT NOW RUNS CLAUDE OPUS 5. NEAR FABLE 5 INTELLIGENCE. HALF THE PRICE. SELF-VERIFIES ITS OWN WORK. AVAILABLE TODAY VIA NOUS PORTAL (20% OFF ALL MODELS). Anthropic shipped Opus 5 on July 24, 2026. same $5/$25 per million tokens as Opus 4.8. but the benchmarks tell a different story. WHAT CHANGED FROM OPUS 4.8: FrontierBench v0.1: Opus 5: 43.3%. Opus 4.8: 18.7%. 2.3x jump on the same test. ARC-AGI-3: Opus 5: 30.2%. 3x better than the next closest model. beat Fable 5 on 8 out of 13 benchmarks. at half the cost ($5/$25 vs $10/$50). same price as Opus 4.8. twice the intelligence. no reason to stay on 4.8. THE SPECS: model ID: claude-opus-5 context: 1M tokens (default and maximum) max output: 128K tokens thinking: on by default effort toggle: low / medium / high per request fast mode: $10/$50, 2.5x faster knowledge cutoff: May 2026 minimum cacheable prompt: 512 tokens (was 1,024) SELF-VERIFICATION (the biggest change): Opus 5 checks its own work automatically. Anthropic says: delete your verification prompts. "include a final verification step" now causes OVER-verification because the model already does it. for Hermes /goal tasks this is a direct upgrade. the judge checks evidence. the model also checks evidence. double layer of verification without extra tokens. EFFORT TOGGLE: low: fast, cheap, routine work. medium: balanced, daily tasks. high: full reasoning, complex problems. set per request. not a global switch. matches Hermes /reasoning command: /reasoning low (routine) /reasoning high (complex) Opus 5 effort toggle + Hermes reasoning control = precise cost management per turn. WHERE OPUS 5 FITS IN HERMES: DAILY DRIVER (replaces Opus 4.8): same price. 2.3x better benchmarks. set as your main model: Desktop app / Dashboard: Models → claude-opus-5 CHIEF OF STAFF: synthesis across multiple agents. reads Kanban, prioritizes, routes tasks. self-verification catches routing errors before they cascade. COMPLEX CODING: SOTA on agentic coding benchmarks. FrontierBench 43.3% = best public model for coding. set as coder profile model. /GOAL TASKS: self-verification + completion contracts = the model proves its work AND double-checks the proof. long-horizon goals finish correctly more often. MoA AGGREGATOR: strongest synthesis model at $5/$25. pair with GPT-5.6 and Grok 4.5 as references. Opus 5 aggregates. best quality at mid-range price. presets: max-quality: reference_models: - provider: openai-codex model: gpt-5.6-sol - provider: xai model: grok-4.5 aggregator: provider: anthropic model: claude-opus-5 COMPUTER USE: near-Fable 5 quality for browser automation. at half the token cost per session. computer_use tasks burn lots of vision tokens. Opus 5 halves that bill vs Fable 5. WHAT TO KEEP OPUS 5 AWAY FROM: cron monitoring: too expensive. use DeepSeek or no_agent mode. sub-agent grunt work: use GPT-5.6 Luna ($1/$6) or DeepSeek. auxiliary tasks: use Gemini Flash. routine web extraction: use a cheap model. Opus 5 is for the turns where quality compounds. planning, synthesis, verification, complex reasoning. budget models handle everything else. NOUS PORTAL: 20% OFF ALL MODELS Nous Portal currently runs a 20% discount on all models including Opus 5. $5/$25 official → $4/$20 through Nous Portal. the cheapest way to run Opus 5 right now. hermes setup --portal select claude-opus-5 as your model. discount applies automatically. Opus 5 replaces Opus 4.8 everywhere. same price. better at everything. no tradeoff. straight upgrade. hermes update /model claude-opus-5
0:10

HERMES AGENT NOW RUNS CLAUDE OPUS 5. NEAR FABLE 5 INTELLIGENCE. HALF THE PRICE. SELF-VERIFIES ITS OWN WORK. AVAILABLE TODAY VIA NOUS PORTAL (20% OFF ALL MODELS). Anthropic shipped Opus 5 on July 24, 2026. same $5/$25 per million tokens as Opus 4.8. but the benchmarks tell a different story. WHAT CHANGED FROM OPUS 4.8: FrontierBench v0.1: Opus 5: 43.3%. Opus 4.8: 18.7%. 2.3x jump on the same test. ARC-AGI-3: Opus 5: 30.2%. 3x better than the next closest model. beat Fable 5 on 8 out of 13 benchmarks. at half the cost ($5/$25 vs $10/$50). same price as Opus 4.8. twice the intelligence. no reason to stay on 4.8. THE SPECS: model ID: claude-opus-5 context: 1M tokens (default and maximum) max output: 128K tokens thinking: on by default effort toggle: low / medium / high per request fast mode: $10/$50, 2.5x faster knowledge cutoff: May 2026 minimum cacheable prompt: 512 tokens (was 1,024) SELF-VERIFICATION (the biggest change): Opus 5 checks its own work automatically. Anthropic says: delete your verification prompts. "include a final verification step" now causes OVER-verification because the model already does it. for Hermes /goal tasks this is a direct upgrade. the judge checks evidence. the model also checks evidence. double layer of verification without extra tokens. EFFORT TOGGLE: low: fast, cheap, routine work. medium: balanced, daily tasks. high: full reasoning, complex problems. set per request. not a global switch. matches Hermes /reasoning command: /reasoning low (routine) /reasoning high (complex) Opus 5 effort toggle + Hermes reasoning control = precise cost management per turn. WHERE OPUS 5 FITS IN HERMES: DAILY DRIVER (replaces Opus 4.8): same price. 2.3x better benchmarks. set as your main model: Desktop app / Dashboard: Models → claude-opus-5 CHIEF OF STAFF: synthesis across multiple agents. reads Kanban, prioritizes, routes tasks. self-verification catches routing errors before they cascade. COMPLEX CODING: SOTA on agentic coding benchmarks. FrontierBench 43.3% = best public model for coding. set as coder profile model. /GOAL TASKS: self-verification + completion contracts = the model proves its work AND double-checks the proof. long-horizon goals finish correctly more often. MoA AGGREGATOR: strongest synthesis model at $5/$25. pair with GPT-5.6 and Grok 4.5 as references. Opus 5 aggregates. best quality at mid-range price. presets: max-quality: reference_models: - provider: openai-codex model: gpt-5.6-sol - provider: xai model: grok-4.5 aggregator: provider: anthropic model: claude-opus-5 COMPUTER USE: near-Fable 5 quality for browser automation. at half the token cost per session. computer_use tasks burn lots of vision tokens. Opus 5 halves that bill vs Fable 5. WHAT TO KEEP OPUS 5 AWAY FROM: cron monitoring: too expensive. use DeepSeek or no_agent mode. sub-agent grunt work: use GPT-5.6 Luna ($1/$6) or DeepSeek. auxiliary tasks: use Gemini Flash. routine web extraction: use a cheap model. Opus 5 is for the turns where quality compounds. planning, synthesis, verification, complex reasoning. budget models handle everything else. NOUS PORTAL: 20% OFF ALL MODELS Nous Portal currently runs a 20% discount on all models including Opus 5. $5/$25 official → $4/$20 through Nous Portal. the cheapest way to run Opus 5 right now. hermes setup --portal select claude-opus-5 as your model. discount applies automatically. Opus 5 replaces Opus 4.8 everywhere. same price. better at everything. no tradeoff. straight upgrade. hermes update /model claude-opus-5

YanXbt

16,744 次观看 • 2 个月前

How to 10x your design with Figma Make ⭐️ I spent 40+ hours testing Figma Make prompts. Most designers waste time with vague prompts and get garbage outputs. Here are the exact prompts and proven workflow that actually work: 1️⃣. Prompt formula: Bad: "Create a dashboard" Good: "Create a SaaS analytics dashboard with: → Left sidebar navigation (240px wide) → Top bar with user profile → 4 metric cards in a grid → Line chart showing revenue trend → Use blue (#2563EB) as primary color" The more you specify = higher quality. 2️⃣ Workflow: Import Your Design System First Before your first prompt: → Go to your main Figma file → Export your component library → Import it into Make → Add this to every prompt: "Use components from [Your Library Name]" Now everything matches your brand automatically. 3️⃣. Prompt for Interactive States: "Create a login form with: → Email and password inputs → Show error state when fields are empty → Disabled button state when form is incomplete → Success message after submission → Add smooth transitions between states" Gets you working prototypes, not static screens. 4️⃣. Advanced Prompts: Data States "Create a user list screen with three states: → Loading (skeleton screens) → Success (populated table with 10 users) → Empty (illustration + 'No users yet' message + 'Add User' CTA)" One prompt = complete UX coverage. 5️⃣. The "Design System Drift” Fix: Notice Make using wrong colors? → Try this Prompt: "Analyze my imported library and list all color tokens, then regenerate using only those exact values" It'll self-correct and stick to your system. 6️⃣. Responsive Design Prompt: "Create a pricing page with 3 tiers. Make it responsive: → Desktop: 3 columns side-by-side → Tablet: 2 columns with 3rd below → Mobile: Stacked vertically → Use Auto Layout for fluid scaling" This gets you mobile + desktop in one shot. 7️⃣. Magic Troubleshoot Prompts: Output looks off? → Try: "Redesign this following Material Design principles" → Or: "Make this follow iOS Human Interface Guidelines" → Or: "Apply Gestalt principles for better visual hierarchy" Give it design frameworks to follow. It works magic. Designers who master prompt engineering in 2026 will ship 10x more than everyone else. P.s. I made a Gameboy for Pokémon. (bookmark this for later)
0:19

How to 10x your design with Figma Make ⭐️ I spent 40+ hours testing Figma Make prompts. Most designers waste time with vague prompts and get garbage outputs. Here are the exact prompts and proven workflow that actually work: 1️⃣. Prompt formula: Bad: "Create a dashboard" Good: "Create a SaaS analytics dashboard with: → Left sidebar navigation (240px wide) → Top bar with user profile → 4 metric cards in a grid → Line chart showing revenue trend → Use blue (#2563EB) as primary color" The more you specify = higher quality. 2️⃣ Workflow: Import Your Design System First Before your first prompt: → Go to your main Figma file → Export your component library → Import it into Make → Add this to every prompt: "Use components from [Your Library Name]" Now everything matches your brand automatically. 3️⃣. Prompt for Interactive States: "Create a login form with: → Email and password inputs → Show error state when fields are empty → Disabled button state when form is incomplete → Success message after submission → Add smooth transitions between states" Gets you working prototypes, not static screens. 4️⃣. Advanced Prompts: Data States "Create a user list screen with three states: → Loading (skeleton screens) → Success (populated table with 10 users) → Empty (illustration + 'No users yet' message + 'Add User' CTA)" One prompt = complete UX coverage. 5️⃣. The "Design System Drift” Fix: Notice Make using wrong colors? → Try this Prompt: "Analyze my imported library and list all color tokens, then regenerate using only those exact values" It'll self-correct and stick to your system. 6️⃣. Responsive Design Prompt: "Create a pricing page with 3 tiers. Make it responsive: → Desktop: 3 columns side-by-side → Tablet: 2 columns with 3rd below → Mobile: Stacked vertically → Use Auto Layout for fluid scaling" This gets you mobile + desktop in one shot. 7️⃣. Magic Troubleshoot Prompts: Output looks off? → Try: "Redesign this following Material Design principles" → Or: "Make this follow iOS Human Interface Guidelines" → Or: "Apply Gestalt principles for better visual hierarchy" Give it design frameworks to follow. It works magic. Designers who master prompt engineering in 2026 will ship 10x more than everyone else. P.s. I made a Gameboy for Pokémon. (bookmark this for later)

Felix Lee

12,706 次观看 • 8 个月前

Idea today. PCB tomorrow. This is GPT-6 Astra working in KiCad, taking a circuit schematic and turning it into an actual PCB layout. It handles things like placing components and routing copper traces which are two of the most time-consuming parts of PCB design. For hardware engineers, this could mean spending less time on repetitive layout work and more time on designing, testing and improving the actual product. AI-assisted hardware design is getting seriously interesting.
0:15

Idea today. PCB tomorrow. This is GPT-6 Astra working in KiCad, taking a circuit schematic and turning it into an actual PCB layout. It handles things like placing components and routing copper traces which are two of the most time-consuming parts of PCB design. For hardware engineers, this could mean spending less time on repetitive layout work and more time on designing, testing and improving the actual product. AI-assisted hardware design is getting seriously interesting.

Praveen Kumar Verma

52,175 次观看 • 25 天前

Fable 5 totally crushed our new contest, but it cost 6x more than Opus 4.8! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics demos Prompts: — A train derailing off a broken bridge into the water — Two cars jumping off ramps and colliding mid-air over a canyon — A monster truck crushing a row of parked cars Outputs: Fable 5: 62,158 tokens, $3.12 GPT 5.5: 37,753 tokens, $1.14 Opus 4.8: 22,280 tokens, $0.56 GLM 5.2: 36,246 tokens, $0.08 Fable 5 did all three scenes at A+. The crashes looked real, things fell and broke the right way, and nothing went through the ground or floated. GPT 5.5 was the closest to Fable. In the Bigfoot show, we think GPT was even a little better. GLM 5.2 did not win any scene, but it was the cheapest by far. Fable is the best pick for quality, but you pay more for it.
0:16

Fable 5 totally crushed our new contest, but it cost 6x more than Opus 4.8! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics demos Prompts: — A train derailing off a broken bridge into the water — Two cars jumping off ramps and colliding mid-air over a canyon — A monster truck crushing a row of parked cars Outputs: Fable 5: 62,158 tokens, $3.12 GPT 5.5: 37,753 tokens, $1.14 Opus 4.8: 22,280 tokens, $0.56 GLM 5.2: 36,246 tokens, $0.08 Fable 5 did all three scenes at A+. The crashes looked real, things fell and broke the right way, and nothing went through the ground or floated. GPT 5.5 was the closest to Fable. In the Bigfoot show, we think GPT was even a little better. GLM 5.2 did not win any scene, but it was the cheapest by far. Fable is the best pick for quality, but you pay more for it.

atomic.chat

2,842,287 次观看 • 3 个月前