Command Code (@CommandCodeAI) header
Command Code (@CommandCodeAI) profile picture

Meet the coding agent built for open models. Command Code. Costco of AI. Model repairs, /design, #1 on leaderboards, world's best $10 GOAT plan w/ $100 usage.

San Francisco, CA
Joined November 2023
Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)Photo by Command Code (@CommandCodeAI)
Introducing the GOAT plan 🐐 Command Code GOAT plan is the best low-cost coding plan on the market today. - $10/mo to get $70 credits. 7x what you pay - 30+ open weight and closed models $1 Go got you started. GOAT gets you shipping. Subscribe — be a GOAT.
257
84
72
1,685
925,836
Command Code retweeted
no GTA day off?? here's the cheat code: /gta
10
8
4
160
11,733
no GTA day off?? here's the cheat code: /gta
10
8
4
160
11,733
GPT-6.2 Astra and GPT-6.2 Sol are now both live in Command Code.
10
3
1
109
10,445
Command Code retweeted
Flint by Recursive is now live & free in Command Code. · 262K token context window · 40B total params, 2.5B active per token · Available all plans and subscribers
2
7
1
119
12,180
Command Code retweeted
Qwen4 Max is now live in Command Code on all plans. - 1M token context window - Multimodal (text, image, video) - 50% less tokens, outperforms Qwen3.8 Max
4
7
2
164
15,152
Command Code retweeted
Hy4 flash is now live in Command Code on all plans. · 1M token context window · 224B total params, 12B active · $0.070 input/ $0.281 output/ $0.003 cache
20
7
2
259
31,984
Command Code retweeted
You can now set the model and reasoning effort for every subagent in Command Code. Keep your main session on one model and run subagents in parallel on another. Try it now!
5
3
1
71
6,438
Flint by Recursive is now live & free in Command Code. · 262K token context window · 40B total params, 2.5B active per token · Available all plans and subscribers
2
7
1
119
12,180
You can now set the model and reasoning effort for every subagent in Command Code. Keep your main session on one model and run subagents in parallel on another. Try it now!
5
3
1
71
6,438
Command Code 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
15,070
Hy4 flash is now live in Command Code on all plans. · 1M token context window · 224B total params, 12B active · $0.070 input/ $0.281 output/ $0.003 cache
20
7
2
259
31,984
Qwen4 Max is now live in Command Code on all plans. - 1M token context window - Multimodal (text, image, video) - 50% less tokens, outperforms Qwen3.8 Max
4
7
2
164
15,152
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
15,070
This week, we shipped several improvements to make the Command Code desktop app faster and more reliable. • Mini chat from anywhere with Shift+Cmd+Space • Manage agents from your menu bar • New themes + VS Code theme import • Edit, reorder, and undo queued messages • Send PR reviews & failed checks to agents • Agents keep running when the window is closed • Smooth performance, even on an 8GB machine • Switch branches without losing work Plus better search, file attachments, usage meters, and much more.
1
1
906