What GPU do you need to run Xiaomi MiMo-V2.6-Flash 309B-A15B?
A new Framework Desktop ($6,799). No priced GPU card holds Xiaomi MiMo-V2.6-Flash 309B-A15B (about 168.9 GB at MXFP4); this system holds it with room for about 996K tokens of context (its full 1M with 8-bit KV cache), at about 24 tok/s estimate. 1 of the 157 GPUs and systems we track fit it.
310.76B parameters, 15B active (mixture-of-experts) (Hugging Face, accessed 2026-10-18). MXFP4 GGUF: 167.37 GB (source). Architecture config (Hugging Face config.json; hybrid_layer_pattern: 9 full + 39 sliding-window (128) layers; SWA K/V dims 192/128 averaged to 160 so the byte count is exact): 9 global attention layers × 4 KV heads × 192 head dim, plus 39 sliding-window layers (128-token window), so about 23 KB of FP16 KV cache per token of context (how we estimate context).
Card: 309B total, 15B activated, 1M context. RL release (the MOPD release of Sep 27 has the same size). Runs in stock llama.cpp (mimo2 architecture; MiMo-V2.6 conversion merged Sep 22, PR #29257). No Q4_K_M found; the size is ggml-org's official MXFP4 GGUF. Release date is the Hugging Face repo creation date.
Other places to get it
Checked Nov 14, 2026. We list only copies we opened and checked. Check file hashes for copies from other uploaders.
- ModelScope: Lab's own account
- Ollama: Cloud only (runs on Ollama's servers)
ollama run mimo-v2.6-flash:cloud
Single GPUs that fit, cheapest first
| Single GPU that fits | VRAM | Speed estimate | Max context estimate | Price |
|---|---|---|---|---|
| Framework Desktop, AMD Ryzen AI Max+ PRO 495 (Strix Halo), 192 GB | 192 GB | ~24 tok/s | 996K · 1M (8-bit) | low $6,799 |
Every device that runs it
Graphics cards, Macs and mini PCs from the wider hardware list behind My rig, smallest usable memory first. Not sure about yours? Check your exact rig.
Show all 4 devices that run it
| Device | Type | Memory | Speed estimate | Max context estimate | Price |
|---|---|---|---|---|---|
| Apple M3 Ultra 256GB | Mac | 256 GB (~192 usable) | ~62 tok/s | 996K · 1M (8-bit) | No price yet |
| Apple M5 Ultra 256GB | Mac | 256 GB (~192 usable) | ~79 tok/s | 996K · 1M (8-bit) | No price yet |
| Framework Desktop, AMD Ryzen AI Max+ PRO 495 (Strix Halo), 192 GB | Mini PC | 192 GB | ~24 tok/s | 996K · 1M (8-bit) | low $6,799 |
| Apple M5 Ultra 512GB | Mac | 512 GB (~384 usable) | ~79 tok/s | 1M (full) | No price yet |
Multi-GPU: cards that need two or more
| Cards | Total VRAM | Speed estimate | Max context estimate |
|---|---|---|---|
| 4 × NVIDIA RTX A6000 | 192 GB | ~73 tok/s | 996K · 1M (8-bit) |
Speed calculator
estimateEstimate = bandwidth ÷ GB read per token (the model's listed GGUF file: Q4_K_M for most models, otherwise the quant named in the list; scaled by active/total parameters for MoE), scaled by a per-platform efficiency and fixed per-token costs calibrated to 45 measured llama.cpp benchmarks (2025–26). Fit = file size + 1.5 GB headroom ≤ total VRAM (on Macs, the part of unified memory macOS lets the GPU use by default), and room for at least 4K tokens of context with an 8-bit KV cache. For a model with no KV cache data, only the file and headroom are checked, so a fit says nothing about context. Multi-GPU: VRAM adds, speed stays one card's (a layer split runs the cards in turn); max context uses the total. Speed is for short context. Max context = what is left after the file and a 1 GB runtime reserve ÷ KV cache bytes per token, default FP16 and 8-bit (q8_0) cache. How speed works · how context works