Skip to content
can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H100 SXM (80GB) run Kimi Linear 48B-A3B (MoE)?

Kimi · MIT · estimated for a 80.0 GB machine at 4K context.

Yes — it runs great.

Runs great

Best fit: Kimi Linear 48B-A3B (MoE) at Q8_0 · 52.3 GB weights · 248–372 tok/s on a NVIDIA H100 SXM (80GB).

Arena Text Overall · style-controlled

UnratedNo exact match

No reviewed exact Arena match

No reviewed exact model/configuration match in the Arena Text Overall style-controlled snapshot. No family or editorial score is substituted. Local quantization quality is unmeasured.

Snapshot published 2026-10-08 · snapshot imported 2026-10-11 UTC · validated last-good data · manual refresh

Arena leaderboard dataset · Arena / lmarena-ai, Arena Leaderboard Dataset (CC BY 4.0). Selected Text Overall style-controlled rows; field names adapted, ratings unchanged. · dataset CC BY 4.0 · revision e16f90232986

Total memory at 4K context
53.1 GB
Weights + KV cache + runtime overhead
Estimated generation
248–372 tok/s
Estimated first token
~0.1s

Every quantization on this machine

On small screens, each quant lists every metric below its name. On wider screens, scroll sideways for total memory, speed and fit. Keyboard: focus the table, then use the arrow keys. Quant names stay visible.

Kimi Linear 48B-A3B (MoE) quantizations on NVIDIA H100 SXM (80GB), estimated at 4K context. Weights and total memory are shown separately.
QuantWeightsTotal memorySpeedFit
Q2_K17.5 GB18.4 GB315–473 tok/sRuns great
Q3_K_M22.7 GB23.5 GB303–455 tok/sRuns great
Q4_K_M30.1 GB30.9 GB287–431 tok/sRuns great
Q5_K_M35.1 GB35.9 GB277–416 tok/sRuns great
Q6_K40.5 GB41.3 GB268–401 tok/sRuns great
Q8_052.3 GB53.1 GB248–372 tok/sRuns great
FP1698.3 GB99.1 GB— not residentWon't fit
Open the full calculator →Everything a NVIDIA H100 SXM (80GB) runsKimi Linear 48B-A3B (MoE) on other machinesWhat to buy to run it →

Ran Kimi Linear 48B-A3B (MoE) on a NVIDIA H100 SXM (80GB)? Add your real numbers →