KymaKyma
MiniMax

MiniMax

MiniMax M2.5

MiniMax's top agentic coding model — 80.2% on SWE-bench — running in Kyma's frontier-open tier with a 100% success rate on production traffic over the measured period.

Modalities

Text → Text

Input

$0.3826 /1M

Output

$1.346 /1M

Cached input

$0.03826 /1M90% off

Context

197K

Speed

medium

Performance

Live production data from real requests on Kyma — not synthetic benchmarks.

Rank

#13

of 87 active models

Tokens served

48.1M

all-time

Success rate

100%

last 7 days

Median throughput46 tok/s
Total requests2,655
Platform share2.0%
Tokens · last 15 daysJul 19 → Aug 2

Pricing

Pay per token. Cached input is billed at 10% of the input rate.

$0.3826 /1M input$1.346 /1M output
Hobby10 req/day · 2K in / 500 out
~$0.43/mo
Production1,000 req/day · 2K in / 500 out
~$43.15/mo
Scale20,000 req/day · 2K in / 500 out
~$863/mo
+ Estimate your workload
1,000
2,000
500
30%

Estimated monthly cost

$36.95

$1.23 / day on MiniMax M2.5

Same workload on:

MiniMax M3$46.21+25%
MiniMax M2.7$48.60+32%

Estimates use list pricing with cached input billed at the 90%-discount rate. Actual bills depend on real token counts — every response includes its exact cost.

When to use MiniMax M2.5

Updated 2026-06-10

Where this model earns its cost — and where it doesn't.

MiniMax M2.5 is MiniMax's frontier-open model built for coding and agentic workflows. Its headline result is an 80.2% score on SWE-bench, the benchmark that measures whether a model can resolve real software issues end to end — which is why it's recommended for coding, agents, and debugging rather than general chat.

On Kyma it has earned real adoption: it ranks 11th of 62 models by tokens served, and coding agents — OpenClaw, Claude Code, and OpenCode — are among its top apps by traffic. Production requests in the measured window succeeded 100% of the time, every call gets automatic failover if a serving path degrades, and prompt caching bills repeated prompt prefixes at 10% of the input rate — which matters for agents that resend long system prompts on every turn.

Capability-wise it covers the full agent loop: a 196K-token context window, function calling, structured outputs, and extended reasoning, with up to 32K tokens of output per request.

Agentic coding

Multi-step software tasks — read the codebase, plan, edit, verify — are exactly what its SWE-bench result measures.

Coding agent backbones

OpenClaw, Claude Code, and OpenCode are already among its top traffic sources on Kyma.

Debugging sessions

Extended reasoning plus a 196K context lets it hold the failing code, logs, and stack traces in one request while it works through the cause.

Structured tool pipelines

Function calling and structured outputs keep multi-step agents emitting parseable, on-schema results.

Long-context code review

The 196K window fits large diffs or several related modules in a single review pass.

Not ideal for: Image inputs (it's text-only — pick a vision model) or latency-critical chat UX, where its medium-speed output (median 22 tokens/sec on Kyma) will feel slower than a fast-tier model.

How it compares

Against the peers people actually weigh it against.

SpecMiniMax M2.5MiniMax M3MiniMax M2.7
Input /1M$0.3826$0.3852$0.405
Output /1M$1.346$1.54$1.62
Context197K1M205K
ToolsYesYesYes
ReasoningYesYesYes
Speedmediummediummedium

Quick start

Up and running in under two minutes.

  1. 1

    Create an API key

    Sign up and grab a key from the dashboard — $0.50 free credit, no card required.

    Get API key →
  2. 2

    Make your first request

    Drop in your key and send a chat completion — fully OpenAI-compatible.

    curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-m2.5",
    "messages": [
      {"role": "user", "content": "Explain prompt caching in one paragraph."}
    ]
  }'
  3. 3

    Stream responses

    Add "stream": true to receive tokens as they arrive.

    curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-m2.5",
    "stream": true,
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

FAQ

Common questions about this model.

What is the context window of MiniMax M2.5?

MiniMax M2.5 has a 197K-token context window — roughly 289 pages of text in a single request.

How much does the MiniMax M2.5 API cost?

$0.3826 per 1M input tokens and $1.346 per 1M output tokens, with cached input at $0.03826/1M — a 90% discount on repeated prompt prefixes. No subscription; you pay only for what you use.

Does MiniMax M2.5 support function calling?

Yes — MiniMax M2.5 supports tool/function calling and structured outputs (JSON mode), so it works with agent frameworks out of the box.

How do I use MiniMax M2.5?

Kyma is OpenAI-compatible: point your SDK's base URL at https://kymaapi.com/v1, use your Kyma API key, and set the model to minimax-m2.5. Signing up is free and includes $0.50 of credit — no card required.

Is MiniMax M2.5 good for agentic coding?

That's its core strength. It scores 80.2% on SWE-bench — resolving real software issues end to end — and on Kyma coding agents like OpenClaw, Claude Code, and OpenCode are among its top apps by traffic. Reasoning, function calling, and structured outputs cover the full agent loop, and production requests in the measured window succeeded 100% of the time.

Is it fast enough for interactive use?

It's a medium-speed model — median 22 tokens/sec on Kyma production traffic. That's a non-issue for agents working through tasks in the background, which is where it shines, but for a chat UI where users watch tokens stream, a fast-tier model will feel snappier.

Why run MiniMax M2.5 through Kyma?

One API key and one OpenAI-compatible endpoint covers this and every other model on the platform. You get automatic failover when a serving path degrades, prompt caching at a 90% discount on repeated prefixes — a big deal for agents resending long system prompts — exact per-request cost in usage.cost, and $0.50 of free credit to start, no card required.

Start with $0.50 free credit — no card required.Create account →

More models by MiniMax

See all 13 →
ModelContextInputOutput
MiniMaxMiniMax M31M$0.3852$1.54
MiniMaxMiniMax Music Pro—$0.21 / song
MiniMaxMiniMax M2.7205K$0.405$1.62
MiniMaxMiniMax Music—$0.045 / song
MiniMaxMiniMax Voice Design—$4.20 / call
MiniMaxHailuo 02 (1080p)—$0.78 / video
MiniMaxHailuo 02 (768p)—$0.42 / video
MiniMaxHailuo 02 (512p)—$0.14 / video