One agent session
1M input tokens and 100K output tokens, no caching.
- Input1M x $0.60 = $0.60
- Output0.1M x $2.40 = $0.24
$0.84per session
MiniMax - Closed weights
MiniMax-M2.5-highspeed pricing and specs
MiniMax-M2.5-highspeed is a closed-weights model from MiniMax that costs $0.60 per 1M input tokens and $2.40 per 1M output tokens on the first-party API, with a 205K-token context window.
Last verified from platform.minimax.io. All sources
- Input / 1M
- $0.60
- Output / 1M
- $2.40
- Context
- 205K tokens
- Weights
- Closed
Pricing per 1M tokens
| Item | Price / 1M tokens |
|---|---|
| Input | $0.60 |
| Output | $2.40 |
| Cache read | $0.03 |
| Cache write (5 minutes) | $0.375 |
Note: A faster serving variant of M2.5 (the launch post calls it M2.5-Lightning: same capability, 100 tps). Marked closed because MiniMax publishes no separate weights repo for the highspeed id; the base M2.5 weights are at huggingface.co/MiniMaxAI/MiniMax-M2.5. Listed under Legacy Models on the pricing page but still in the Available Models list. Being retired: unavailable to new users from 2026-10-19 and retired on 2026-11-18; re-check or remove after 2026-11-18. release date is from the M2.5 launch post (2026.2.12), which announced both versions. Text and tool-call input only. max output is the documented max_tokens ceiling for non-M3 models and includes thinking tokens.
Specs
- Context window
- 204,800 tokens
- Max output
- 204,800 tokens
- Release date
- Provider
- MiniMax
- API model id
MiniMax-M2.5-highspeed- Vercel AI Gateway id
minimax/minimax-m2.5-highspeed- Weights
- Closed weights
What MiniMax-M2.5-highspeed costs in practice
Worked from the standard list rates above, with the math shown so you can swap in your own token counts.
A chat app
10,000 requests a day at 2K input and 500 output tokens each.
- Input20M x $0.60 = $12.00
- Output5M x $2.40 = $12.00
$24.00per day
$720.00 per 30-day month
The same session at a 90% cache hit
90% of the input is read from the prompt cache. The one-time cache write premium is not included.
- Cached input0.9M x $0.03 = $0.027
- Uncached input0.1M x $0.60 = $0.06
- Output0.1M x $2.40 = $0.24
$0.327per session
Compare with
| Model | Input / 1M | Output / 1M |
|---|---|---|
| MiniMax-M2.5-highspeed (this page) | $0.60 | $2.40 |
| MiniMax-M2.7-highspeedMiniMax | $0.60 | $2.40 |
| Qwen3.5-27BAlibaba Cloud (Qwen) | $0.30 | $2.40 |
| Qwen3.5-PlusAlibaba Cloud (Qwen) | $0.40 | $2.40 |
| Amazon Nova 2 LiteAmazon | $0.30 | $2.50 |
| MiniMax-M2.5MiniMax | $0.30 | $1.20 |
| MiniMax-M2.7MiniMax | $0.30 | $1.20 |
| MiniMax-M3MiniMax | $0.30 | $1.20 |
Related reading
- MiniMax H3: An Omni-Modal Video Model With Native Audio, 2K Output, and Open Weights ComingMiniMax H3 turns text, image, video and audio input into 2K video with native stereo sound at 0.80 CNY per second, with open weights promised to follow.
- MiniMax M2.5 for Developers: The Anthropic-Compatible Budget Frontier ModelMiniMax M2.5 hits 80.2% on SWE-bench Verified and plugs into the Anthropic SDK with two environment variables. Here is what you need to know before switching.
FAQ
How much does MiniMax-M2.5-highspeed cost per million tokens?
MiniMax-M2.5-highspeed costs $0.60 per 1M input tokens and $2.40 per 1M output tokens on MiniMax's first-party API, per platform.minimax.io on October 17, 2026. Per 1M tokens, cache reads cost $0.03.
What is MiniMax-M2.5-highspeed's context window?
MiniMax-M2.5-highspeed has a 204,800-token context window and returns up to 204,800 output tokens per response.
Are MiniMax-M2.5-highspeed's weights open?
No. MiniMax-M2.5-highspeed is a closed-weights model. You call it through MiniMax's API as MiniMax-M2.5-highspeed or through Vercel AI Gateway as minimax/minimax-m2.5-highspeed.
When was MiniMax-M2.5-highspeed released?
MiniMax released MiniMax-M2.5-highspeed on February 12, 2026.
Sources
Last verified from platform.minimax.io.
- https://platform.minimax.io/docs/guides/pricing
- https://platform.minimax.io/docs/pricing/overview
- https://platform.minimax.io/docs/guides/text-generation
- https://platform.minimax.io/docs/faq/models
- https://platform.minimax.io/docs/api-reference/text-anthropic-api
- https://platform.minimax.io/docs/api-reference/text-chat-anthropic
- https://platform.minimax.io/docs/release-notes/models
- https://www.minimax.io/news/minimax-m25
Newsletter
Get model price changes as they ship
One email per week. New model releases, price cuts and what they mean for your bill.