MiMo-V2-Flash
Where it runs 8K context · ≥ 5 t/s floor
| Setup | Best plan | Memory | Generate |
|---|---|---|---|
| 8 GB GPU + 32 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 12 GB GPU + 32 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 16 GB GPU + 64 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 24 GB GPU + 64 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 32 GB GPU + 128 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 2× 24 GB GPUs + 128 GB RAM | No usable expert/non-expert split: the config does not describe it, or it disagrees with the publisher active count (see model page) | ||
| 96 GB workstation GPU + 256 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU layers · 26/48 layers on GPU | KV unknown | ~15–21 t/s Estimated extrapolated |
| Mac, 24 GB | Does not fit | ||
| Mac, 48 GB | Does not fit | ||
| Mac, 128 GB | Does not fit | ||
| Mac, 512 GB | Q8_0llama.cpp MetalUnified memory | KV unknown | ~9.7–28 t/s Estimated extrapolated |
| DGX Spark 128 GB | Does not fit | ||
| Ryzen AI Max+ 395, 128 GB | Does not fit | ||
| CPU only, 64 GB DDR5-5600 dual channel | Does not fit | ||
Same plan engine as the planner, with its default reserves and RAM. How plans are chosen.
Files exact Hugging Face bytes
| Artifact | Size | Files | Revision | Source |
|---|---|---|---|---|
| Q8_0 bartowski/XiaomiMiMo_MiMo-V2-Flash-GGUF | 306 GiB | 9 | 3047e46bcf | HF files |
| Q5_K_M bartowski/XiaomiMiMo_MiMo-V2-Flash-GGUF | 204 GiB | 6 | 3047e46bcf | HF files |
| Q4_K_M bartowski/XiaomiMiMo_MiMo-V2-Flash-GGUF | 174 GiB | 5 | 3047e46bcf | HF files |
| MLX 4-bit 4-bit mlx-community/MiMo-V2-Flash-4bit | 162 GiB | 36 | 65d2673628 | HF files |
| Native safetensors mixed/custom native; quant_method=fp8; stored tensor dtypes= XiaomiMiMo/MiMo-V2-Flash | 292 GiB | 145 | 1afd314a24 | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, MLX 8-bit, EXL3, AWQ, GPTQ. Complete selected file sets; projectors, MTP/draft and lookup files are listed separately and budgeted only by plans that use them. File bytes are not allocated memory.
Quality
No Artificial Analysis evaluation mapped to this checkpoint.
Arena text rating
Unknown
Publisher-reported benchmarks
None extracted from the model card.
AA and Arena describe the hosted model in the stated mode, not each quantized file; scores are never inherited by base, pruned, abliterated or quantized variants and never averaged together.
How to run it
No model-specific recipe in the research. The planner binds the backend’s documented command template (llama.cpp, Ollama, LM Studio, MLX, vLLM) to your plan: open a plan’s details.
Recipes are source examples (settings are the publisher’s or author’s, not measured optima) and were not executed by LocalFit. A downloadable file does not prove your backend build supports the architecture.
Reported speeds 0
No reported measurement yet. Submit one.