impulse

MiniMax M2.7

Open weights

MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.

API id MiniMax-M2.7Released 18 Mar 2026 (Mar 2026)
Input / 1M
$0.21
at GMI Cloud
Output / 1M
$0.84
at GMI Cloud
Context window
197K
tokens
Max output
1M
tokens per reply
Hosts
10
10 listings
Released
Mar 2026
18 Mar 2026

Pricing

List prices at GMI Cloud, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.21$0.84
Cached input
80% off input
$0.042

What it costs a month

At GMI Cloud list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$111
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$90.72
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$161
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$706
Use your own numbers

Where to run it

10 hosts serve MiniMax M2.7. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
minimax-m2-7
$0.15 / $0.60
−29% vs headline
minimax-m2.7
$0.255 / $0.55
−11% vs headline
GMI CloudInference
MiniMax-M2.7
$0.21 / $0.84
DeepInfraInference
MiniMax-M2.7
$0.25 / $1.00
+19% vs headline
AIHubMixGateway
minimax-m2.7
$0.296 / $1.18
+41% vs headline
Together AIInference
MiniMax-M2.7
$0.30 / $1.20
+43% vs headline
Novita AIInference
minimax-m2.7
$0.30 / $1.20
+43% vs headline
minimax-m2.7
$0.405 / $1.62
+93% vs headline
SambaNovaInference
MiniMax-M2.7
$0.60 / $2.40
+186% vs headline
Fireworks AIInference
minimax-m2p7
-

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: Apache-2.0
Runs in about 548.9 GB of GPU memory
Runtimes: SGLang, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • EmpirioLabs AI
    minimax-m2-7
    $0.15 / $0.60
    29% less
  • Pinstripes
    minimax-m2.7
    $0.255 / $0.55
    11% less

Smaller MiniMax models

No cheaper MiniMax model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    46% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    46% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    29% less

Open-weight picks

  • Muse Glimmer 30B
    Meta · on EmpirioLabs AI
    $0.20 / $0.80
    Has every key capability
    5% less
  • Mistral Small 2603
    Mistral AI · on Mistral AI
    $0.15 / $0.60
    Has every key capability
    29% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    31% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    51% less

Model line

Newer and older versions, at each lab's own price.

  1. May 2026MiniMax M3Latest$0.30 / $1.20
  2. Mar 2026MiniMax M2.7This page$0.21 / $0.84
  3. Feb 2026MiniMax M2.5$0.30 / $1.20
  4. UndatedMiniMax M2.1$0.30 / $1.20
  5. UndatedMiniMax M2$0.30 / $1.20

Call it

Using GMI Cloud's API.

Python
from litellm import completion

response = completion(
    model="gmi/MiniMax-M2.7",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does MiniMax M2.7 cost?

On GMI Cloud, MiniMax M2.7 costs $0.21 per million input tokens and $0.84 per million output tokens. Cached input costs $0.042 per million tokens. The cheapest host is EmpirioLabs AI, at $0.15 per million input tokens and $0.60 per million output tokens.

Which APIs offer MiniMax M2.7?

MiniMax M2.7 runs on 10 APIs: EmpirioLabs AI, Pinstripes, GMI Cloud, DeepInfra, AIHubMix, Together AI, Novita AI, Kyma API, SambaNova and Fireworks AI.

What is MiniMax M2.7's context window?

197K tokens, with up to 1M tokens in one reply.

Is MiniMax M2.7 being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of MiniMax M2.7?

Yes. MiniMax M3 is the newest model in this line, released on 31 May 2026.

What are cheaper alternatives to MiniMax M2.7?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (46% less), GPT-6 Luna on OpenAI (46% less) and Muse Glimmer 30B on EmpirioLabs AI (5% less). MiniMax M2.7 itself costs 29% less on EmpirioLabs AI. Savings are on a mix of 3 input tokens to 1 output token.

When was MiniMax M2.7 released?

MiniMax released it on 18 Mar 2026.

What is MiniMax M2.7's API model id?

Use MiniMax-M2.7 with GMI Cloud. Other hosts use their own ids: minimax-m2-7 on EmpirioLabs AI, minimax-m2.7 on Pinstripes, minimax-m2.7 on AIHubMix, minimax-m2.7 on Novita AI.

Sources

Raw catalog entry for MiniMax-M2.7 on GMI Cloud
{
  "cache_read_input_token_cost": 4.2e-8,
  "input_cost_per_token": 2.1e-7,
  "litellm_provider": "gmi",
  "max_input_tokens": 196608,
  "mode": "chat",
  "output_cost_per_token": 8.4e-7,
  "source": "https://ie-api.gmicloud.ai/v1/artifact/get_public_artifacts"
}