Visual Synthesizer - Daniel Joy Grimes

🎨 Artist by day 🌙 Creative Coder by night

Computational Artist circa 2016 | PromptEngineering | LLMs | OpenSource | Linux | Computational Muralism

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

Last year, I created 5 murals for a really unique exhibition. I used neural networks to recreate a Chief, Chief Pamblo, that was never photographed. Its one of the first examples of AI being used for cultural restoration. Then I used my custom computational muralism pipelines to generate the Murals. The murals were over a billion pixels.

The artworks below were created in collaboration with the Nevada City Rancheria (NCR) Nisenan Tribe guided by CHIRP's Visibility Through Art (VTA) Initiative. This intentional and informed collaboration includes hours of consultation with the Tribe to best address their visibility needs. VTA is intended to generate Cultural awareness and build an understanding of the Nisenan through art.
https://t.co/AovrtXt6qq
#MilesDearling #Nisenan #CulturalRestoration #ChiefPamblo #Pamblo #visionaryart

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

Full methodology and caveats: https://t.co/5L6RdhFasE

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

MiniMax-M2.7 NVFP4: 127.7 tok/s C=1, 2800 peak C=128

2x RTX PRO 6000 Blackwell, SGLang b12x 0.8.3, TP=2.
Prefill 17,286 tok/s @ 8K → 9,908 @ 128K.

No spec decode (no drafter yet — when one ships, low-C numbers should jump).

Thanks @lukedobl for the quant + kernels.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

**All data is public:**
- Results & methodology: [https://t.co/PN956OdwuD](https://t.co/0hICWMhEt6)
- Raw benchmark JSONs: [https://t.co/wrCFL6erbr](https://t.co/Sx7nyOxzoW)
- 3-run verification data: [run1](https://t.co/KYrC17VAVe), [run2](https://t.co/kbRK2PWI4y),
[run3](https://t.co/yGIkfyXEwT)

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

198 tok/s on Qwen3.5-122B with 2 GPUs
148 tok/s on M2.5 (229B)
79 tok/s on Qwen3.5 397B (full GGUF)

~15% faster than a WRX90 Threadripper build on the same stack.

Stack:
122B NVFP4 — 198 tok/s (SGLang b12x + NEXTN)
27B FP8 — 170 tok/s (DFlash drafter)
229B NVFP4 — 148 tok/s (vLLM b12x)
397B GGUF — 79 tok/s (llama.cpp)

Hardware:
ASRock Rack B650D4U (AM5)
EPYC 4564P (AM5)
PLX switch
2x RTX PRO 6000

3-run verified: 197 / 200 / 198
Reproducible. Link in comments.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

Local AI in 2026 vs 2020.

Same energy. ~198x the intelligence.

Cables got worse (waiting on shorter cables). Benchmarks got better.

198 tok/s | Qwen3.5-122B | 2 GPUs | no NVLink
148 tok/s | MiniMax M2.5 229B | vLLM b12x
79 tok/s | Qwen3.5 397B | GGUF llama.cpp

Build:
$2K AM5 ASRock Rack B650
$3K PCIe switch
2x RTX PRO 6000 Blackwell

Result:
Beats $33K Threadripper builds by ~18% on the same stack

Open-weight + local is not just speed.

It is:
• privacy
• data sovereignty
• control

Some work should never leave the machine.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@Teknium Congratulations! I'm loving Hermes Agent.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

Local AI in 2026 vs 2020.

Same energy. ~198x the intelligence.

Cables got worse (waiting on shorter cables). Benchmarks got better.

198 tok/s | Qwen3.5-122B | 2 GPUs | no NVLink
148 tok/s | MiniMax M2.5 229B | vLLM b12x
79 tok/s | Qwen3.5 397B | GGUF llama.cpp

Build:
$2K AM5 ASRock Rack B650
$3K PCIe switch
2x RTX PRO 6000 Blackwell

Result:
Beats $33K Threadripper builds by ~18% on the same stack

Open-weight + local is not just speed.

It is:
• privacy
• data sovereignty
• control

Some work should never leave the machine.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@JeffSynthesized @Kling_ai wow- it looks rad!

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

yeah its rough, i decided just to stay on my old gen4 threadripper. it already had 256gb ram. its a tiny bit slower for AI inference on blackwell 6000s- but paying for rd-dimm right now is crazy. thats like the cost of another blackwell 6000.

also playing around with AM5 servers soon. only 1 gen5pcie slot, but it might not matter for my workflows.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@sagharborcap not sure about that. i learned calculus and linear algebra using Khan to build AI.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@gen_ericai i was using MJ before it was even named, helped build datasets they used for aesthetic fine tuning- but I haven't touched it in years. Its not a professional art tool. Switch to open models and black box API's.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

We’ll see about the timeline. At a certain point it may not matter. For example, with image generation I can already produce commercial-quality work using NanoBanana Pro or open-source models such as Flux2 or Qwen. Raw image quality is comparable. The main difference is workflow control and platform restrictions (censorship). Open-source models perform far better when a project requires a high degree of control or multiple revision cycles. For simple probabilistic generation with minimal creative direction, black-box systems are sufficient. But for professional work, I need deterministic, iterative control loops.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@PurzBeats yeah! So true. im really interested in deterministic art systems with generative assistance. constantly inspired by the creativity of the community.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@jordandchesney sadly its a cultural issue and a legal issue. Open source will catch up in a few months though so seedance wont really matter for long

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@hellorob its great until its not. ever actually used it on commercial work? the censorship makes it hard to take black box models seriously.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@AIWarper unless you make memes, even closed source seems like a waste of time.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

to be fair ive nuked my own inbox a couple times

[Summer Yue @summeryue0]

Nothing humbles you like telling your OpenClaw “confirm before acting” and watching it speedrun deleting your inbox. I couldn’t stop it from my phone. I had to RUN to my Mac mini like I was defusing a bomb.

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@jtydhr88 🔥

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

@peterom 🔥🔥🔥 best discord Ive been in since the early Eleutherai 2020 colab days. I learn so much. thank you

[Visual Synthesizer - Daniel Joy Grimes @visual_synth]

very true. ive always found the obsession with realism boring artistically. i think it gimped the art form. replicating a camera or older art styles has always been low hanging fruit and not artistically interesting. Looking forward to watching your video. been a while since i played with AD