The AI arena is free today

Open Superagent
API Provider10 active models1 organization

MiniMax: API pricing, speed & models

MiniMax hosts 10 active AI models, with input pricing from $0.30 per 1M tokens, with median throughput of 106 characters/sec, and P95 time to first token of 3.02s, with 99% success rate over 7 days. Compare MiniMax's API speed, pricing, and reliability against other inference providers.

AudioVideoFirst-party only
Median throughput106 c/s
Median TTFT793ms
P95 TTFT3.02s
Success rate (7d)99%
From$0.30 /M tok

Catalog

Type
Price
20 models
Model
MiniMax M3cache pricing
MiniMax M3cache pricing
MiniMax M3cache pricing
MiniMax M2.5cache pricing · 100 QPS
MiniMax M2.5cache pricing · 100 QPS
At a glance

MiniMaxpricing, performance & catalog

The citable facts about MiniMax's 10 models — sourced from provider APIs and refreshed continuously.

Lowest price
MiniMax M3 at $0.300 per 1M input tokens
Highest median throughput
MiniMax M2.7 at 229 chars/s
Lowest median TTFT
MiniMax M3 at 0.79s
Largest context
MiniMax M3 at 1.0M tokens
Catalog
10 active models from 1 organization

FAQ

Common questions about MiniMax.

What is MiniMax?

MiniMax is an API provider that hosts large language models. Active models: 10; From (input): $0.30 / 1M tok; Median throughput: 106 c/s; Median TTFT: 793ms; Success rate (7d): 99%.

How many models does MiniMax offer?

MiniMax currently serves 10 active models out of 10 historical offerings on LLM Stats.

What is MiniMax's API pricing?

MiniMax input pricing starts from $0.30 per 1M tokens, with the most expensive offering at $0.3 per 1M tokens. See the Pricing tab above for the full per-model breakdown.

How fast is MiniMax?

MiniMax delivers a median throughput of 106 characters per second and P95 time to first token of 3.02s across its model catalog. See the Models tab for per-model throughput and TTFT breakdowns.

Is MiniMax reliable?

MiniMax has a 99% success rate across 395 API calls in the last 7 days, with a 1% error rate.

Does MiniMax support function calling?

Yes. 8 of 10 models on MiniMax support function calling (tool use). The Capabilities tab lists which specific models accept tool definitions.

Does MiniMax support JSON mode and structured output?

Yes. 8 of 10 models on MiniMax support structured output (JSON mode / schema-constrained generation). The Capabilities tab shows which specific models accept response_format or json_schema parameters.

Does MiniMax offer batch inference?

Yes. 8 of 10 models on MiniMax support batch inference for cheaper, asynchronous workloads.

Does MiniMax support multimodal models?

Yes. MiniMax's catalog includes 4 vision-capable, 8 audio, and 4 video models. See the Models and Capabilities tabs for the full per-model breakdown.

Whose models does MiniMax host?

MiniMax hosts models from MiniMax. See the Models tab for the full catalog grouped by creator.

How do I start using MiniMax?

Sign up at https://platform.minimax.io to get an API key, then call MiniMax's API directly from your application. Most clients work out of the box by pointing the OpenAI SDK at MiniMax's base URL with your key. Use the Models and Pricing tabs above to pick the right model for your latency, cost, and context-window requirements.