GPU rankings
MiniMax M3 428B

Cheapest GPU for MiniMax M3

B300 is the cheapest at $0.056 per million tokens, 16% below B200.

Measured on a single-turn chat workload (8k input / 1k output) at a matched interactivity target of 50 tokens/s per user. Hardware without a measurement at this operating point is not ranked.

Cheapest GPU for MiniMax M3: live benchmark ranking
RankGPUTokens/s per GPU$ / 1M tokensPrecisionEngine
1B30011,209$0.056fp4vLLM
2B2007,193$0.067fp8vLLM
3GB200 NVL724,780$0.11fp8Dynamo vLLM
4GB300 NVL725,837$0.11fp8Dynamo vLLM
5MI355X3,681$0.11fp4vLLM

Optimizing for speed instead of cost? See the fastest GPU for MiniMax M3

Methodology

Every number on this page is a measurement, not a spec-sheet estimate. The InferenceX fleet serves MiniMax M3 on real hardware with community serving engines, sweeping concurrency to trace each platform's throughput-versus-interactivity frontier.

Platforms are then read at the same operating point (50 tokens/s per user) so the comparison is iso-interactivity: a GPU cannot win by quoting throughput at an unusably slow per-user speed. Cost converts measured throughput to $ per million total tokens using $/GPU/hr rates from the SemiAnalysis AI Cloud TCO model.

The derivation is shared with the InferenceX overview leaderboard, and results re-run continuously, so this ranking updates as new engine releases and configs land.

Frequently asked questions

What is the cheapest GPU to run MiniMax M3?
As of the latest benchmark runs, B300 is cheapest at $0.056 per million total tokens on a single-turn chat workload (8k input / 1k output), at hyperscaler $/GPU/hr pricing and a matched interactivity target of 50 tokens/s per user.
How is cost per million tokens calculated?
Measured throughput per GPU at the 50 tokens/s per user tier is converted to $ per million total (input plus output) tokens using hyperscaler $/GPU/hr rates from the SemiAnalysis AI Cloud TCO model. Slower interactivity targets or cheaper rental tiers change the absolute numbers but rarely the order.
How often is this ranking updated?
Benchmarks re-run continuously on the InferenceX cluster fleet; the newest result feeding this ranking landed on 2026-08-21. The page always renders the latest data.

Explore the data