MiniMaxCheap API

Cheap MiniMax H3 API

Run MiniMax's official MiniMax H3 from $0.044/s — 45% below the official $0.08/s. Pay per second, no subscription, failed generations refunded.

From: $0.044/s; 768P
Official: $0.08/s; MiniMax
You save: up to 45%
100 × 5 s clips: $22.00; official $40.00

Try in Studio Call the API

Price check

MiniMax H3: our price vs the official price

Every MiniMax H3 tier, priced live from our catalog next to MiniMax's official list price.

MiniMax H3 price per second: official list price and Cheaper Gen. price for every tier
Tier	MiniMax official	Cheaper Gen.	You save	5 s clip
768P per second	~~$0.08~~/s	$0.044/s	−45%	$0.22$0.40
2K per second	~~$0.13~~/s	$0.0715/s	−45%	$0.36$0.65

Official price: MiniMax international official price ($0.08 / $0.13 per second; reference-video seconds billed at the output rate, reference audio free). Verified October 16, 2026. Reference-video seconds are billed at the output rate; reference audio is free.

At scale

What MiniMax H3 costs at volume

Totals for common batch sizes at today's per-second prices.

MiniMax H3 cost examples at volume
Batch	Tier	Official	Cheaper Gen.	You keep
100 × 5 s clips	768P	$40.00	$22.00	$18.00
100 × 5 s clips	2K	$65.00	$35.75	$29.25
60 minutes of footage	768P	$288.00	$158.40	$129.60

Every tier and a calculator on the MiniMax H3 pricing page

Why it costs less

Same official MiniMax H3, lower price

The model and its output are the same; the cost of serving it is lower. How our inference works.

Official partnerships

Direct partnerships with the official model labs give us their lowest partner and channel pricing on official models like MiniMax H3.

Our own inference engine

For open-source models, our own inference engine — GPU scheduling, warm pools and per-model tuning — lowers the cost of every output further.

Automatic routing

Every MiniMax H3 request is routed to the most cost-effective healthy channel, with live health checks and failover.

Spend less

Ways to pay even less for MiniMax H3

Draft at 768P, deliver at 2K

A second of 768P costs $0.044 — 62% of the 2K rate ($0.0715/s). Iterate on prompts and camera moves cheaply, then render the final cut with the same prompt.

Test with 5-second clips

You pay per second of output (5–15 s per clip). A 5-second 768P test costs $0.22; lengthen only the takes you keep.

Failed generations cost nothing

The price is held when you submit and refunded in full if a generation fails or you cancel it while it is queued. You only pay for outputs you get.

Volume pricing below these rates

Committed monthly volume unlocks rates below the public price list, plus invoicing and dedicated capacity.

Talk to sales

Cheaper alternatives

Cheaper alternatives to MiniMax H3

Other official video models with an overlapping task, cheapest first. Price per second at each model's lowest tier.

PixVerse V6PixVerse · 360p · up to 20% off$0.0176/s

Seedance 2.0 MiniByteDance · 480p · up to 43% off$0.02/s

Veo 3.1 LiteGoogle · 720p · up to 59% off$0.0206/s

LTX-2.5Lightricks · 720p · up to 76% off$0.022/s

All MiniMax H3 alternatives compared

Start generating with MiniMax H3

Try MiniMax H3 in the Studio, or call it from your code with one API key. A 5-second 768P clip costs $0.22 and failed generations are refunded.

Open in Studio API reference

curl -X POST https://cheapergen.com/api/v1/run/minimax/h3/text-to-video \
  -H "Authorization: Bearer $CHEAPERGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "…"}'

FAQ

Cheap MiniMax H3 API questions

Our prices on this page are live; MiniMax's official prices were verified October 16, 2026.

MiniMax H3 starts at $0.044/s (768P) against MiniMax's official $0.08/s. Across all 2 pricing tiers you pay 45% less than the official list price. Official prices were verified October 16, 2026; our prices on this page are live.

Yes. MiniMax H3 is developed by MiniMax. Cheaper Gen. provides API access to the official model through official upstream endpoints — the same model and output quality, at a lower price. We don't own the model.

Three things: direct partnerships with the official model labs give us their lowest partner and channel pricing; our own inference engine for open-source models (GPU scheduling, warm pools, per-model tuning) lowers costs further; and automatic routing sends every request to the most cost-effective healthy channel, with failover. Most of the saving is passed on in the price.

100 five-second clips at 768P cost $22.00 here, versus $40.00 at the official price — $18.00 saved. Longer clips and higher resolutions scale per second; the pricing page has every tier.

There is no subscription and no monthly fee. Cheaper Gen. is prepaid: add credits from $10, spend them on any model in the API or the Studio, and failed generations are refunded automatically. The quote endpoint (POST /v1/quote/{endpoint}) is free, so you can price any request before you run it.

MiniMax H3 is developed by MiniMax. Cheaper Gen. provides API access through official upstream endpoints and does not own the model. Trademarks belong to their respective owners.