Submit
MiniMaxOpen-weight·1000K context·— params

MiniMax M3minimax/minimax-m3

MiniMax-M3 accepts text, image and video input with text output. The official API documents a 1M context. Its base USD price applies to standard requests with up to 512K input tokens; longer inputs and priority admission use separate rate bands. Check the exact configuration before budgeting.

Cheapest blended:$0.53 / 1M tokenson MiniMax · 1 provider listed

Official sources checked October 2, 2026. Evaluation suggestions below are not benchmark results.

Match the displayed price to the request configuration

The matrix shows the standard lower-input band, not every request configuration. Long inputs and priority admission have distinct rates. Use the official rate card or the updated MiniMax API pricing article to calculate a mixed workload by configuration.

The current official API lists a 1M context. No maximum output ceiling has been verified in the sources reviewed here, so this page removes the earlier unsupported numeric limit. Confirm the accepted output budget on your endpoint rather than copying an intermediary's limit.

Keep M3 and M3.1 Preview access separate

M3 remains the documented pay-as-you-go option. M3.1-Flash-Preview is currently restricted to M Plan and MiniMax Code. A new preview name does not establish a retirement or automatic rerouting of M3.

Before switching, compare credential type, reasoning handling and billing configuration. Use the same task set and keep failures in the cost ledger. The pricing review establishes documented rates, not an independent performance result.

Official source 1 →Official source 2 →Official source 3 →

Pricing across providers

Sort by:
ProviderInput /1MOutput /1MBlended /1MLatency p50FormatFreshnessAction
MiniMax
MiniMax-M3

USD standard rate for requests with ≤512K input tokens. Longer inputs and priority admission have separate rates. See the current source and MiniMax pricing guide.

$0.3
Cached $0.06
$1.2$0.525—OpenAI-compatibleNeeds 24d agoTry →

Affiliate disclosure: We may earn a commission from qualified signups. Pricing independence is enforced at the data layer — see our Editorial Independence Policy.

Works with

Point any of these clients at a hosting's base URL — they all speak at least one of this model's endpoint protocols (OPENAI_COMPATIBLE).

Capabilities

Code samples

Example using MiniMax — the cheapest hosting for this model as of last verification. Swap base_url and model to use a different provider from the matrix above.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.minimax.io/v1",
)

response = client.chat.completions.create(
    model="MiniMax-M3",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Technical specs

Context
1000K
Max output
—
Parameters
—
Release
—
Training cutoff
—
License
—

Similar models

Compare with

  • MiniMax M3 vs DeepSeek V4.1 Flash
    Comparison planned — not yet published
  • MiniMax M3 vs ERNIE 4.5 VL 424B A47B
    Comparison planned — not yet published
  • MiniMax M3 vs GLM 4.5V
    Comparison planned — not yet published

Frequently asked

How much does MiniMax M3 cost?+
The cheapest public hosting is $0.53 per 1M blended tokens on MiniMax. 1 total providers are listed above with per-input / per-output / cached pricing.
How do I access MiniMax M3 from outside China?+
Availability, account verification and payment methods vary by provider and region. Check the official documentation and the selected provider's access requirements before integrating. A directory listing does not guarantee global API access.
Is MiniMax M3 open-source? Can I fine-tune it?+
MiniMax M3 has published weights. License: see the official repository. Review the linked repository and license for deployment, fine-tuning and commercial-use conditions.
Is MiniMax M3 OpenAI-compatible?+
At least one listed hosting exposes an OpenAI-compatible API, so you can point an existing openai SDK client at the Provider's base_url and use the Provider's model name. See the Code Samples above for a copy-pasteable example.
What's the maximum context window for MiniMax M3?+
The model supports up to 1,000,000 tokens of context (input + output). Some hosted versions may impose a smaller limit — check the "Context" column in the pricing matrix for each provider.