What GPU do you need to run MiniMax-M3.1-Flash?
No single GPU. MiniMax-M3.1-Flash needs about 259.5 GB at Q4, more than any GPU, system or build we price.
427.05B parameters, 23B active (mixture-of-experts) (Hugging Face, accessed 2026-10-23). Q4_K_M GGUF: 258.01 GB (source). Architecture config (Hugging Face config.json; sparse_attention_freq is 1 on all 60 layers, including the 3 leading dense-MLP layers (moe_layer_freq). llama.cpp minimax-m3 sets leading_dense_block_count from those leading zeros and allocates the indexer cache only on the other 57 layers: 1 KV head, K and V both sparse_index_dim 128, same element type as the main cache (src/llama-kv-cache-msa.cpp). The graph writes indexer K only; V is still allocated): 60 attention layers × 4 KV heads × 128 head dim, plus an indexer cache on 57 sparse layers (1×128 K and V), so about 149 KB of FP16 KV cache per token of context (how we estimate context).
Card: approximately 428B total, 23B activated per token, text/image/video in, 1M context, MiniMax Sparse Attention on every layer. Parameter count is the checkpoint's tensor shapes with NVFP4 expert tensors (weight_packed, U8) counted at 2 parameters per byte and quant-scale tensors excluded: 427.05B. Hugging Face's safetensors total is 220.49B stored elements, because those experts are packed. The weights ship already quantized (NVFP4 routed experts, MXFP8 attention) plus a separate DSpark draft that is not in the GGUF. Q4_K_M is AtomicChat's plain quant for stock llama.cpp (minimax-m3, no patch): 258.01 GB, the sum of the six Q4_K_M shards. The same repo has Q3_K_M at 203.49 GB and Q2_K at 155.91 GB (its own page: minimax-m3.1-flash-q2-k). Fit ignores the 1.73 GB vision projector. Commercial use of the MiniMax Community License needs attribution and, above $20M yearly revenue, written authorization. Release date is the Hugging Face repo creation date. Announced at https://x.com/MiniMax_AI/status/2113436864904610008.
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 minimax-m3.1-flash:cloud
No single GPU in our database fits it. Use multiple GPUs (see calculator).
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 1 devices that run it
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