Model by MiniMax

MiniMax M3.1 Flash

MiniMax M3.1 Flash is the fast, low-cost M3.1 model from MiniMax.

Checked 2026-11-05

MiniMax M3.1 Flash is the fast, low-cost M3.1 model from MiniMax. OpenRouter says it improves on M3 across several benchmarks, most of all in complex agentic work, coding and cybersecurity. It was listed on 19 October 2026. The context window is 1,048,576 tokens, it takes text, image and video, and it returns text. On 5 November 2026 the listing price was $0.15 per million input tokens and $0.60 per million output, with GMICloud lowest at $0.1125 and $0.45. The weights are public on Hugging Face under a licence labelled minimax-community. We could not read the licence text, so check its terms before shipping a product. The model card says about 428B total and 23B active parameters, though the same page shows a 220B figure in its file metadata and does not explain the gap. It suits cheap agent loops and high-volume coding calls. The limit is that output is text only. Reach it through MiniMax, OpenRouter, or other hosts. Figures come from OpenRouter and the Hugging Face card; we have not run it.

Mentioned here

  • MiniMax M3

    MiniMax M3 is an open-weights model released 2026-05-31, with a 1,048,576-token context and text, image and video input.

    • Open weights

    MiniMax, 1M context, $0.30 in / 1M

  • Kimi K3

    Kimi K3 is an open-weights model from Moonshot AI, released 2026-07-16, with a 1,048,576-token context and text, image and video input.

    • Open weights

    Moonshot AI, 1M context, $0.19 in / 1M

  • DeepSeek V4.1 Flash 1017

    DeepSeek V4.1 Flash 1017 is a low-cost open-weights DeepSeek model released 2026-10-17.

    • Open weights

    DeepSeek, 1M context, $0.15 in / 1M

  • GLM 5.3 Flash

    GLM 5.3 Flash is a very low-cost open model from Z.ai, released 2026-08-26 under the MIT licence, with a 1,048,576-token context and text, image and video input.

    • Open weights

    Zhipu AI (Z.ai), 1M context, $0.15 in / 1M

  • OpenRouter

    OpenRouter routes more than 600 models through one API key.

    • Model gateway (router)
    • OpenAI compatible

  • Hugging Face Inference

    Hugging Face Inference Endpoints deploys a model from the Hub onto dedicated, autoscaling hardware.

    • Model / inference API
    • OpenAI compatible

Questions

What is the context window of MiniMax M3.1 Flash?

1,048,576 tokens, with up to 524,288 tokens of output.

How much does MiniMax M3.1 Flash cost?

$0.15 per million input tokens and $0.6 per million output tokens, checked 5 November 2026.

Can I run MiniMax M3.1 Flash on my own hardware?

Yes, the weights are published under the Open-weights model licence. They are on huggingface.co.

Who makes MiniMax M3.1 Flash?

MiniMax makes it, and released it on 19 October 2026.

Reviews

Loading reviews…

Write a review

Your rating

Up to 1000 characters. Stars alone are enough.

Similar

  • MiniMax M2.5

    MiniMax M2.5 is an open-weights coding model from MiniMax, suited to teams that want cheap agent runs and can accept that it has been superseded.

    • Open weights

    MiniMax, 200K context, $0.27 in / 1M

  • MiniMax M2.7

    MiniMax M2.7 is an agent-oriented text model from MiniMax, and it suits teams that want a mid-priced model for tool-heavy coding work.

    • Open weights

    MiniMax, 197K context, $0.21 in / 1M