---
type: "Article"
title: "MiMo-V2.6-Flash-MOPD 309B-A15B with vLLM on 2× DGX Spark"
description: "MiMo-V2.6-Flash-MOPD 309B-A15B with vLLM on 2× DGX Spark: recipe by myllmbox with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/myllmbox--mimo-v26-flash-mopd-2x-dgx-spark"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-11-17T02:08:15.637Z"
status: "stable"
---

# MiMo-V2.6-Flash-MOPD 309B-A15B with vLLM on 2× DGX Spark

MiMo-V2.6-Flash-MOPD 309B-A15B with vLLM on 2× DGX Spark: 49 tok/s according to github.com (dataset as of Nov 3, 2026).

- Creator: [myllmbox](https://www.contextstudios.ai/local-ai/creators/myllmbox.md)
- Engine: vLLM 0.30.1rc1 (Nightly ddd6fbca, Image myllmbox/mimo-v26-flash-cluster-vllm:v2: mbx-Loader, Split-KV-Attention beim Spec-Decoding), TP=2 über RDMA, MTP K=3 (3 trainierte Draft-Layer)
- Quantization: Offizieller Checkpoint wie veröffentlicht (config.json: quant_method fp8, store_dtype mxfp4); KV-Pool 16 GB/Box ≈ 1,22M Token; Marlin MoE
- Model family: MiMo-V2.6-Flash-MOPD
- Hardware: dgx-spark x2
- Context: 300000

## All numbers

- 49 tok/s (Peak; Peak, 1 request, code-heavy spec prompt (SQL/TypeScript), thinking off, with MTP K=3) — [source](https://github.com/myllmbox/mimo-v26-flash-cluster-recipe#measured-performance-this-exact-stack-2-dgx-spark-rdma-mtp-k3-vmcompaction_proactiveness0)
- decode_tps: 49 (1 request) — [source](https://github.com/myllmbox/mimo-v26-flash-cluster-recipe#measured-performance-this-exact-stack-2-dgx-spark-rdma-mtp-k3-vmcompaction_proactiveness0)
- prefill_tps: 1947 (1 request) — [source](https://github.com/myllmbox/mimo-v26-flash-cluster-recipe#long-context-one-request)
- ttft_ms: 690 (1 request) — [source](https://github.com/myllmbox/mimo-v26-flash-cluster-recipe#long-context-one-request)

## Weights

- [XiaomiMiMo/MiMo-V2.6-Flash-MOPD](https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-MOPD) (main, mit)

[https://github.com/myllmbox/mimo-v26-flash-cluster-recipe](https://github.com/myllmbox/mimo-v26-flash-cluster-recipe)

## Related

- [Local AI on your hardware.](https://www.contextstudios.ai/local-ai.md)
