Mia's banner
Mia's profile picture

Mia

@MiaAI_lab • 43,931 subscribers

Building with AI & LLMs // Running frontier open models locally

Shorts

Here's one way to get the most from Grok 4.5 ✨ Switching effort to "low" gives strong results on most tasks, near-zero quality loss, big usage savings — plus it's the fastest.

Here's one way to get the most from Grok 4.5 ✨ Switching effort to "low" gives strong results on most tasks, near-zero quality loss, big usage savings — plus it's the fastest.

1,312,268 views

People realizing they can run Opus-4.6 level intelligence on their laptop with Qwen3.8-27B. No subscription. No limits. Completely private.

People realizing they can run Opus-4.6 level intelligence on their laptop with Qwen3.8-27B. No subscription. No limits. Completely private.

167,988 views

My RTX 5090 is officially doing his part for local AI

My RTX 5090 is officially doing his part for local AI

23,663 views

DeepSeek v4.1 Flash on 3x DGX Sparks

DeepSeek v4.1 Flash on 3x DGX Sparks

30,649 views

One task, two models. GLM 5.3 Flash running slower at 25 tok/s prose Qwen3.8 Flash running faster at 47 tok/s prose Who will finish the task first?

One task, two models. GLM 5.3 Flash running slower at 25 tok/s prose Qwen3.8 Flash running faster at 47 tok/s prose Who will finish the task first?

27,481 views

Until there’s an official Grok Build app, you can still use it from your iPhone or Android with Termius (or any other SSH client). Just install Tailscale on your server, connect to the Tailscale IP in Termius, and run Grok Build from the command line as usual 👇

Until there’s an official Grok Build app, you can still use it from your iPhone or Android with Termius (or any other SSH client). Just install Tailscale on your server, connect to the Tailscale IP in Termius, and run Grok Build from the command line as usual 👇

15,627 views

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks runs so well, feels too good to be true. SOTA AI at home.
0:28
MiaAI_lab's profile picture

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks runs so well, feels too good to be true. SOTA AI at home.

Mia

53,382 views • 4 days ago

DeepSeek V4.1 Flash 1017 on TensorFold ⚡️ - 16 concurrent session - Each with full native 1M context - Non-greedy, thinking on. OFFICIAL WEIGHTS. Recipe dropping very soon!
0:53
MiaAI_lab's profile picture

DeepSeek V4.1 Flash 1017 on TensorFold ⚡️ - 16 concurrent session - Each with full native 1M context - Non-greedy, thinking on. OFFICIAL WEIGHTS. Recipe dropping very soon!

Mia

21,711 views • 6 days ago

Whoever made this is a genius 😂
1:45
MiaAI_lab's profile picture

Whoever made this is a genius 😂

Mia

246,498 views • 2 months ago

You'd be surprised which Claude model won this. I tested Opus 5.5, Fable 5.1, Sonnet 5.5 & Haiku 5.5. All on High effort. Who built the best Gummy Bears? 🥇 Sonnet 5.5 did the BEST by far, it's not even close. As expected, Haiku performed the worst. Opus and Fable are close, so it's a matter of personal taste. I recently claimed that Sonnet 5.5 on high is better than Opus 5.5 on med, but in here, every model is on high, and Sonnet beat them all! Live demo link below.
0:55
MiaAI_lab's profile picture

You'd be surprised which Claude model won this. I tested Opus 5.5, Fable 5.1, Sonnet 5.5 & Haiku 5.5. All on High effort. Who built the best Gummy Bears? 🥇 Sonnet 5.5 did the BEST by far, it's not even close. As expected, Haiku performed the worst. Opus and Fable are close, so it's a matter of personal taste. I recently claimed that Sonnet 5.5 on high is better than Opus 5.5 on med, but in here, every model is on high, and Sonnet beat them all! Live demo link below.

Mia

21,165 views • 6 days ago

DeepSeek v4.1 Flash via API is almost TOO fast 🤯
0:21
MiaAI_lab's profile picture

DeepSeek v4.1 Flash via API is almost TOO fast 🤯

Mia

139,414 views • 1 month ago

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks ⚡️ - Official weights on TensorFold - 8 concurrent streams with thinking on
0:35
MiaAI_lab's profile picture

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks ⚡️ - Official weights on TensorFold - 8 concurrent streams with thinking on

Mia

22,649 views • 8 days ago

Kimi K3.1 Pro does Gummy Bears. Max effort. For a new, top of the line 2.8T param model that puts itself in direct comparison with Opus 5.5, this is only an 'ok' result imo. I'm not impressed. To build it, it used 2x 5h usage limits on a $39 plan 😅
0:48
MiaAI_lab's profile picture

Kimi K3.1 Pro does Gummy Bears. Max effort. For a new, top of the line 2.8T param model that puts itself in direct comparison with Opus 5.5, this is only an 'ok' result imo. I'm not impressed. To build it, it used 2x 5h usage limits on a $39 plan 😅

Mia

19,095 views • 7 days ago

GLM 5.3 Flash vs DeepSeek V4.1 Flash 1017 Can't believe we're abusing gummy bears to test models, but here we are! Both models were run locally. - GLM 5.3 Flash: my own EXL3 quant (link below) - DeepSeek V4.1 Flash 1017: official quant - Same harness.Both on Max effort. Who won?
0:33
MiaAI_lab's profile picture

GLM 5.3 Flash vs DeepSeek V4.1 Flash 1017 Can't believe we're abusing gummy bears to test models, but here we are! Both models were run locally. - GLM 5.3 Flash: my own EXL3 quant (link below) - DeepSeek V4.1 Flash 1017: official quant - Same harness.Both on Max effort. Who won?

Mia

32,512 views • 12 days ago

This is why DGX Spark prices have sky rocketed
0:28
MiaAI_lab's profile picture

This is why DGX Spark prices have sky rocketed

Mia

43,094 views • 23 days ago

Full GLM 5.3 running on 3x DGX Sparks with TensorFold Currently 24-28 tok/s on prose
0:31
MiaAI_lab's profile picture

Full GLM 5.3 running on 3x DGX Sparks with TensorFold Currently 24-28 tok/s on prose

Mia

49,756 views • 28 days ago

That Argon gummy bear video turned out to be a FAKE. I ran the prompt myself, and the results from Gemini 4 Argon is MUCH better. Opus is still better, but Argon is not bad at all. Full prompt below.
0:30
MiaAI_lab's profile picture

That Argon gummy bear video turned out to be a FAKE. I ran the prompt myself, and the results from Gemini 4 Argon is MUCH better. Opus is still better, but Argon is not bad at all. Full prompt below.

Mia

27,506 views • 17 days ago

Highly recommend to watch the whole thing and pay a special attention to the top-right of the screen.
0:23
MiaAI_lab's profile picture

Highly recommend to watch the whole thing and pay a special attention to the top-right of the screen.

Mia

46,133 views • 1 month ago

DeepSeek v4.1 Flash running on 2x DGX Sparks
0:35
MiaAI_lab's profile picture

DeepSeek v4.1 Flash running on 2x DGX Sparks

Mia

59,572 views • 1 month ago

Opus 5.5 built me a Fortnite replica in one prompt
1:55
MiaAI_lab's profile picture

Opus 5.5 built me a Fortnite replica in one prompt

Mia

44,111 views • 1 month ago

Qwen3.8 Flash I really like this model and find it great for frontend. It performs much better than Qwen3.8 27B.
0:22
MiaAI_lab's profile picture

Qwen3.8 Flash I really like this model and find it great for frontend. It performs much better than Qwen3.8 27B.

Mia

61,327 views • 1 month ago

- GLM 5.3 Flash EXL3 - 2x DGX Sparks - TensorFold - 4 concurrent prose streams Pay attention to tok/s and TTFT on each stream. Buttery smooth experience!
0:37
MiaAI_lab's profile picture

- GLM 5.3 Flash EXL3 - 2x DGX Sparks - TensorFold - 4 concurrent prose streams Pay attention to tok/s and TTFT on each stream. Buttery smooth experience!

Mia

32,203 views • 1 month ago

Jumpgate - hyperspace jump DS4.1F-1017 vs GLM 5.3 Flash vs Fable 5.1 Who won imo: 🥇 Fable 5.1 🥈 DeepSeek V4.1 Flash 1017 🥉 GLM 5.3 Flash You can run these by yourself here: Prompt below.
0:24
MiaAI_lab's profile picture

Jumpgate - hyperspace jump DS4.1F-1017 vs GLM 5.3 Flash vs Fable 5.1 Who won imo: 🥇 Fable 5.1 🥈 DeepSeek V4.1 Flash 1017 🥉 GLM 5.3 Flash You can run these by yourself here: Prompt below.

Mia

12,939 views • 11 days ago

I ran this locally with GLM-5.3 Flash EXL3 on my DGX Sparks, and I think it beats both Fable 5.1 and GPT-6 Astra. The crazy part: it was faster to make create this on my sparks than it was with either Fable 5.1 or Astra. Live link here:
0:25
MiaAI_lab's profile picture

I ran this locally with GLM-5.3 Flash EXL3 on my DGX Sparks, and I think it beats both Fable 5.1 and GPT-6 Astra. The crazy part: it was faster to make create this on my sparks than it was with either Fable 5.1 or Astra. Live link here:

Mia

48,031 views • 1 month ago

Check out how DeepSeek v4.1 Flash performs on 2x DGX Sparks after the update. Smooth experience!
0:36
MiaAI_lab's profile picture

Check out how DeepSeek v4.1 Flash performs on 2x DGX Sparks after the update. Smooth experience!

Mia

39,502 views • 1 month ago

Models now include a proper filtering per device type, and consists 16 different recipes! Includes DGX Sparks, RTX 5090, RTX 6000 Pro, and 16-24 GB GPUs. Let me know what you think!
0:35
MiaAI_lab's profile picture

Models now include a proper filtering per device type, and consists 16 different recipes! Includes DGX Sparks, RTX 5090, RTX 6000 Pro, and 16-24 GB GPUs. Let me know what you think!

Mia

30,322 views • 1 month ago