MiniMax M2.5
minimaxai/minimax-m2.5
cross-platform full-stack agentic dev.
cmd --model minimaxai/minimax-m2.5
Intelligence index
33.7
Output speed
91.1 tok/s
Input
$0.30 /M
Output
$1.20 /M
Cache read
$0.03 /M
Agent-loop cost
$0.11 /M in
Context window
200K tokens
Released
February 12, 2026
Modalities
→
vs. the lineup
MiniMax M2.5 beside its stablemates and nearest rivals. The ◆ marks the best value in each column across every row shown.
pin a rival:
| Model | Intelligence | Coding | Speed | Input $/M | Output $/M | Blended $/M | Context |
|---|---|---|---|---|---|---|---|
| MiniMax M3 | 44.4◆ | 58.6◆ | 90.3◆ | $0.30◆ | $1.20◆ | $0.52◆ | 1M◆ |
| Tencent Hy3 | 41.2◆ | 58.8◆ | 65.2◆ | $0.14◆ | $0.58◆ | $0.25◆ | 262K◆ |
| MiniMax M2.7 | 38.1◆ | 52.6◆ | 59.8◆ | $0.30◆ | $1.20◆ | $0.52◆ | 200K◆ |
| Kimi K2.5 | 35.4◆ | 46.8◆ | 41.2◆ | $0.60◆ | $3◆ | $1.20◆ | 256K◆ |
| MiniMax M2.5 ◆ | 33.7◆ | —◆ | 91.1◆ | $0.30◆ | $1.20◆ | $0.52◆ | 200K◆ |
| Step 3.7 Flash | 30.3◆ | 39.6◆ | 399.5◆ | $0.20◆ | $1.15◆ | $0.44◆ | 256K◆ |
| DeepSeek V4 Propinned | 44.3◆ | 59.4◆ | 70.9◆ | $0.43◆ | $0.87◆ | $0.54◆ | 1M◆ |
| DeepSeek V4 Flashpinned | 40.3◆ | 56.2◆ | 122◆ | $0.14◆ | $0.28◆ | $0.18◆ | 1M◆ |
coding performance
The Intelligence Index and its sub-scores, ranked against every scored model in the catalog. A metric that has not been measured for MiniMax M2.5 has been left empty.
IFBench
71.6
instruction following
Intelligence Index
33.7
#36 of 40 scored
Long-context reasoning
66
reasoning across a long context
SciCode
42.6
scientific coding
GPQA Diamond
84.8
graduate-level QA
usage calculator
How far a month of credits goes on MiniMax M2.5.
Input tokensfresh prompt
800Output tokensmodel reply
180Cache read tokensre-read context
50Kcost / request $0.0020 · in $0.30 · out $1.20 · cache $0.03 per M
fresh input 12%output 11%cache reads 77%
Requests / 30 days
5.1K
$10 credits ÷ $0.0020 per request
~1.0K quick fixes~204 bug fixes~34 feature PRs
what real work costs
Real coding tasks priced end to end on MiniMax M2.5, from a quick lookup to a full-repo agent run.
One agent task
$0.03
180K in at 75% cache hit, 12K out
What you pay
$0.17 /M
all-in across every token that task touched
Sticker input
$0.30 /M
cache reads bill at $0.03 /M instead
| Task | Tokens in · out | MiniMax M2.5 | Tencent Hy3 | Claude Haiku 4.5 |
|---|---|---|---|---|
| Quick lookup / one-liner | 8K · 1K | $0.0017 | $0.0009 | $0.0065 |
| Review a 500-line PR | 60K · 4K | $0.01 | $0.0063 | $0.04 |
| Fix a bug (agent loop) | 180K · 12K | $0.03 | $0.02 | $0.12 |
| Refactor a module | 320K · 20K | $0.06 | $0.03 | $0.21 |
| Full-repo agent run | 900K · 45K | $0.13 | $0.08 | $0.49 |
frequently asked
What is MiniMax M2.5 best for?+
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.
How much does MiniMax M2.5 cost?+
$0.30/M input and $1.20/M output, cache reads $0.03/M. In an agent loop most input is cache-read, so the effective input rate is about $0.11/M.Which plan do I need?+
Available on Go and above.
How do I switch to it?+
Run
cmd --model minimaxai/minimax-m2.5, or type /model in a session and pick it. You can switch mid-session without losing context.Ship code that matches your taste
Command Code is the AI coding agent that continuously learns your taste. Start for $1.
commandcode.ai/models/minimax-m2-5