Providers /Minimax
Inference provider

Minimax

Models
11
Regions
00

Our take

Editorial by LLMap · updated Jul 29, 2026

Minimax is a niche provider offering a tightly-curated family of proprietary models with moderate throughput. Its catalogue is small and focused, but significant gaps remain in compliance, geography and data-handling disclosures.

Use this when you specifically need Minimax's proprietary model family and cannot source it elsewhere. Pick it for latency-tolerant workloads where 26 to 65 tokens per second is acceptable. Skip it if you need verified compliance credentials, a known headquarters location, or clear data-residency and prompt-handling policies.

Strengths
  • Highest-throughput offer reaches 65 tokens per second on minimax-minimax-m2-1.
  • Low entry price point available: the cheapest input tier in the family is well under half the price of the highest.
Trade-offs
  • Extensive compliance and operational data gaps: headquarters country, EU endpoint, prompt training policy, SOC 2, and zero-retention option are all unverified in our data.
  • Throughput is unmeasured for 3 of 11 offers.
  • Output pricing varies by more than double within the same family, and duplicate model names with different rates create confusion.
01

Privacy & data handling

Yes
No
No answer on record

The colour says whether the answer favours you, not whether it is a yes — not training on your prompts earns a green cross, no zero-retention option earns a red one.

Trains on your prompts
No
Logs prompts
Yes
Offered
No answer on record
02

Models & pricing

Something wrong on this page? Tell us