Naymur Rahman (@naymur_dev) header
Naymur Rahman (@naymur_dev) profile picture

Design & Code @CommandCodeAI ✹ Building a Frontend ecosystem https://nitter.cf/t.co/zjrS1rOibJ

Joined January 2023
Naymur Rahman retweeted
Don't use Sloppy Tabs. Deslop it using /design from Command Code.
Your badge looks Sloppy, Just Deslop It.
1
60
Naymur Rahman retweeted
I benchmarked the new Qwen4-Max against HY4-Flash on 3D Visual. Same prompt, same iterations, both finished it using /design from Command Code. HY4-Flash is the cheapest option and still produced a noticeably better visual result.
1
1
35
3,079
Naymur Rahman retweeted
Your badge looks Sloppy, Just Deslop It.
1
1
9
519
Naymur Rahman retweeted
Last week in Command Code, a quick recap. · GLM-5.3 usage increased to $70 in GOAT · Muse Ember (2x usage on GOAT) · Inkling 2 Small (FREE model) · Gemini 4 Carbon & Claude Fable 5.2 · Command Code Ambassadors Program · Customizing your keyboard shortcuts · Multiple improvements in desktop app · Command Code at first SGLang Summit
12
8
1
209
14,957
Naymur Rahman retweeted
Deslop card on any vibecoded app
2
1
91
Naymur Rahman retweeted
Deslop is really easy now. Just use /design from Command Code. you don't need anything else.
1
126
Naymur Rahman retweeted
Muse Ember surprised me with the output 👀 A lightweight model created this 3D scene using the /design feature from Command Code. Muse Ember nailed it in one shot, while GPT-6-Luna needed two prompts to get it perfect.
Muse Ember is now live in Command Code on all plans. · $10/mo GOAT gets $20 usage · 1M token context window · GPT-6 Luna & Haiku 5.5 level price Try it out!
7
9
1
63
16,227
Naymur Rahman retweeted
Command Code GOAT plan gets you 10x value. For $10/mo get $100 usage across DeepSeek V4.1 Flash 1017, GLM-5.3 Flash, and more. Try it now. Be GOATED!
17
9
1
237
15,879
Naymur Rahman retweeted
Introducing Cut for macOS. Cut ⌘X and Paste ⌘V files in Mac. Finally. Open source. MIT. No network.
7
3
33
2,470
Naymur Rahman retweeted
Ling 3.1 Flash 1019 is FREE in Command Code for the next 4 days. It's the latest upgrade to Ling 3.1 Flash, an open-weight model for advanced reasoning and agentic workflows. - 1M token context window - 560B total params, 25B active Available on all plans & subscriptions.
16
12
7
274
31,169
Naymur Rahman retweeted
If you’re using DeepSeek harness with Command Code’s provider API, we are shipping several updates to make your experience better and faster. DeepSeek harness needs more than a standard chat completions endpoint. I’m excited to pass along our learnings and improvements we have made here without y’all having to do anything on your end. LFG!!
18
7
174
11,135
Naymur Rahman retweeted
Earlier today we detected and stopped a security incident linked to a large-scale distillation attack targeting Anthropic and OpenAI models. As a precaution, we've revoked all API keys. To keep using Command Code, log in to Studio again to generate a new key. So far, we've found no evidence that any users were impacted. We're working with Cloudflare, Anthropic, and OpenAI to investigate. We appreciate your patience and trust. 💜
118
27
58
571
119,471
Naymur Rahman retweeted
I compared Mimo-V2.6-Flash, Grok-4.7, and DeepSeek-V4.1-Flash. Same prompt with the /design command from Command Code. Reviewed gameplay UX/UI and cost: - Mimo-V2.6-Flash: 9/10 · $0.005 · one-shot playable game with the cheapest price - DSV4.1-Flash: 9/10 · $0.0089 · one-shot with good UI/UX & playable - Grok 4.7: 8/10 · $0.20 · expensive, and I had to iterate to fix the bird
4
3
55
9,393
Naymur Rahman retweeted
Extremely simple to understand, $1 plan has no API ❌ $10 plan includes API ✅ I don't know why it's really hard for someone to understand. Stop abusing good things 😊
1
127
Naymur Rahman retweeted
makes me happy to see our users enjoying command code, faster inference on deepseek has literally cut my work hours in half. think i'm shipping 2-3x more prs a day now. and we're growing so fast, winning hearts, minds, and support of the dev community! 50t tokens now! 🐐
26
4
135
5,456
Naymur Rahman retweeted
I compared DeepSeek-v4.1-Flash against Opus 5 and GPT-5.6 Sol. Same prompt with the /design command from Command Code. Opus 5 is extremely good at UI/UX, but DSV4.1-Flash was almost 90% cheaper than GPT-5.6 Sol and Opus 5💀 What do you think? is DSV4.1-Flash really that good?
10
7
2
129
23,525
Naymur Rahman retweeted
DSV4.1-Flash is good and fast. Compared DSV4.1 with GLM-5.3 Flash and Gemini 3.8 Flash. - GLM-5.3 Flash ($0.0043): was low-cost and strong on design tasks. - Gemini 3.8 Flash ($0.205): was pricier and needed many iterations to finish it. GLM-5.3 Flash wins this benchmark.
🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient. 🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models. 1/6
2
1
1
22
3,371
Naymur Rahman retweeted
GPT-6 Astra created this entire game in one-shot 🔥
1
3
388
Naymur Rahman retweeted
I ran flappybench with Fable 5.1 vs GLM 5.3 and Kimi K3 3 models. Same prompts. All using Command Code /design feature. Kimi K3 and GLM 5.3 beat Fable 5.1 on both design and price. GLM 5.3 stands out, clean, smooth visuals, a strong UI, and the best bird design (better than Kimi). OS works really well with GOATed harness.
1
1
14
3,085
Naymur Rahman retweeted
2
1
2
133