Kimi-K3.1-Pro VRAM Requirements & Cheapest Live GPU Rates

Exact VRAM math for Kimi-K3.1-Pro at FP16, FP8 and INT4 across 8K to 131K context, matched with live RunPod and Vast.ai rates for the cheapest fitting GPU rigs.

Architecture truth card — Kimi-K3.1-Pro

Exact safetensors params2,780,110,363,392 (2780.1B)
Hidden layers93
KV heads (GQA)1
Head dim288
Data snapshot (UTC)2026-11-18T18:29:13+00:00
Snapshot SHA-256d89025de2d15…

← Open the interactive VRAM & cost calculator on the homepage

Exact VRAM matrix — 3 precisions × 3 context lengths

Precision8192 tokens ctx32768 tokens ctx131072 tokens ctx
FP16/BF16 16-bit5178.36 / 0.82 / 260.42
Total 5439.59 GiB
5178.36 / 3.27 / 260.42
Total 5442.05 GiB
5178.36 / 13.08 / 260.42
Total 5451.86 GiB
FP8/INT8 8-bit2589.18 / 0.82 / 130.96
Total 2720.96 GiB
2589.18 / 3.27 / 130.96
Total 2723.41 GiB
2589.18 / 13.08 / 130.96
Total 2733.22 GiB
INT4/AWQ 4.5-bit1456.41 / 0.82 / 74.32
Total 1531.55 GiB
1456.41 / 3.27 / 74.32
Total 1534.00 GiB
1456.41 / 13.08 / 74.32
Total 1543.81 GiB

Each cell: Weights GiB / GQA KV Cache GiB / Runtime Overhead GiB; bold line = Total VRAM GiB. Concurrency = 1, FP16 KV cache (2 bytes), overhead = 1.5 + 0.05 × weights GiB.

Terminal speed-math — no install needed

Re-run this exact sizing from your own terminal:

curl -fsSL https://gpuvram.pages.dev/vram-cli.py | python3 - kimi-k3-1-pro --ctx 32768

Live cheapest GPU configurations

ScenarioCheapest fitting rig (top 3)Total VRAMLowest $/hr730 h/moDeploy
INT4 @ 32K ctx8x MI300X 192GB1536 GiB$4.00$2920.00Deploy on RunPodCheck Vast.ai Verified
INT4 @ 32K ctx8x B300 288GB2304 GiB$55.52$40529.60Deploy on RunPodCheck Vast.ai Verified

What over-provisioning costs you

Kimi-K3.1-Pro at INT4 / 32K context fits the cheapest rig (8x MI300X 192GB) for $2920.00/mo (730 h). Scaling the same INT4 workload to 131K context pushes the KV cache past that rig: the cheapest 131K-capable rig (8x B300 288GB) bills $40529.60/mo — a concrete $40529.60 instead of $2920.00 gap you pay monthly for headroom the 32K rig physically lacks.