(Corrected we are not sure about the model size) Hi everyone, Architectural Foundations & Design MiniMax M3 is a native multimodal model pre-trained on over 100 trillion tokens of interleaved multimedia data, allowing it to excel at complex tasks like agentic coding (scoring 59.0% on SWE-bench Pr…

I wonder if nvidia planned for it. Otherwise a double nic is a waste. But makes sense for a ring topology

I am running this recipe on 3 nodes RIGHT NOW. It “works” but there is a lot of issues. I curently got it at 375K context. The main issue right now is it keeps getting stuck in loops when I ask a question the reasoning just LOOPS. 10tok/s I guess I can take a swing at the TP=3

How are you running it? Via RAY? But VLLM does not take tp=3? SGLang? Jeeez.. I can’t make a financial responsible reasoning to buy 2 more sparks + router + cable as returns are negligible for me, but 1 more + 1 basic cable for extra GPU and 128 GB? Hell yes if it works..

Unlikely, however one thing I will do over the next few weeks is to see if my theory of a fully meshed 4 node cluster is physically possible with this hardware. Should have the required cable for that proof of concept at some stage in the next two weeks (3-5 wees lead time sucks).

SGLang, PP=3 (pipeline parallel) — not Ray, not TP. You’re right that vLLM won’t do TP=3 on M3: the head dims don’t divide by 3 (64 attn / 4 KV heads), so tensor-parallel across 3 GPUs isn’t possible out of the box. That’s why I’m on sglang’s pipeline-parallel for now. NCCL runs over plain TCP betw…

Please keep us informed. This is incredible opportunity to get the hardware that can be used before price becomes astronomical.

For tp=3 you need to run the b12x developer’s VLLM fork which integrates it. I’ve got some more cables coming so will try it as well next week. It’s one of the release branches in here: GitHub - local-inference-lab/vllm: A high-throughput and memory-efficient inference and serving engine for LLMs · …

[image] p33zy: For tp=3 you need to run the b12x developer’s VLLM fork which integrates it. I’ve got some more cables coming so will try it as well next week. It’s one of the release branches in here: GitHub - local-inference-lab/vllm: A high-throughput and memory-efficient inference and serving…

Any idea if SGlang works well enough with GGUF to use the unsloth GGUF’s of M3?

tried IQ4_XS - result = 10.7 tok/sec

MiniMax M3 : NVFP4 for Quad DGX Spark

Accelerated Computing DGX Spark / GB10 User Forum DGX Spark / GB10

tonyd615 June 14, 2026, 10:50pm 113

2 node spark vs 3 or 4 node spark

Topic		Replies	Views
MiniMax M3 NVFP4 and NVFP4 REAP 50 for 4x & 2x DGX Sparks DGX Spark / GB10 Projects	53	4447	July 2, 2026
MiniMax-M3 (428B MoE + vision) at ~14–15 tok/s on 2× DGX Spark — EAGLE3 speculative decoding is the unlock DGX Spark / GB10 Projects llama , deepseek	2	832	July 4, 2026
Minimax3 on 2 nodes decode ~10.7 tok/s, 4bits DGX Spark / GB10 llama	26	1696	June 21, 2026
MiniMax M2.5 released (not available on HuggingFace as of now) -- is DGX Spark ready? DGX Spark / GB10	92	7122	April 12, 2026
MiniMax M2.7 NFVP4 Recipe & Benchmarks DGX Spark / GB10 llama	126	13568	August 20, 2026
MiniMax-M3-W4A16-GPTQ 2xGB10 Deployment (36 t/s) (fp8, nvfp4, KVarN, EAGLE-3) DGX Spark / GB10 agentic-ai	25	3020	August 17, 2026
MiniMax-M3-AWQ on 4× GB10, fp8 KV, 262k context, adaptive reasoning, ~30 tok/s DGX Spark / GB10 llama	17	1342	July 21, 2026
DGX Spark performance DGX Spark / GB10	49	7305	February 13, 2026
MiniMax 2.5 REAP - NVFP4 on single DGX Spark DGX Spark / GB10	25	3455	April 1, 2026
Best 2026 model for agentic work on a 2-node Spark cluster? DGX Spark / GB10 agentic-ai , deepseek , openclaw	33	6211	July 1, 2026

MiniMax M3 : NVFP4 for Quad DGX Spark

Related topics