Context Window

Compare

Google Gemma 4 31b vs Kimi K2 Thinking Maas

This page is context-first: how much text each model can take in one request. Full specs adds capabilities and limits; the pricing matrix below is only about $/million tokens from hosts that list both models.

Google

Model

Google Gemma 4 31b

Image inputTool calling

Context window

256K

256,000 tokens · ~192K words

Model page
Moonshot

Model

Kimi K2 Thinking Maas

Tool calling

Context window

256K

256,000 tokens · ~192K words

Model page

Context window · side by side

Bar length is relative to the larger of the two windows (100% = max of this pair). This is not pricing.

Google Gemma 4 31b256K
Kimi K2 Thinking Maas256K

Same context window size for both models.

Token calculator →Memory savings →

Google Gemma 4 31b and Kimi K2 Thinking Maas have identical context windows (256K tokens). Google Gemma 4 31b is 76% cheaper on input.

Quick verdicts

Short takeaways — validate with your own workloads.

  • RAG / high-volume retrieval

    Use Google Gemma 4 31b. Input tokens are 76% cheaper — critical when sending large retrieved contexts.

Full specs

Context, output, capabilities, and dates. Green highlights the favorable value where we compute a winner.

SpecGoogle Gemma 4 31bKimi K2 Thinking Maas
Context window256,000 tokens (256K)256,000 tokens (256K)
Max output tokens256,000 tokens (256K)256,000 tokens (256K)
Speed tierFastDeep
VisionYesNo
Function callingYesYes
Extended thinkingYesNo
Prompt cachingNoNo
Batch APINoNo
Release dateN/AN/A

Pricing matrix

Dollar rates only: hosts that list both models, per 1M tokens. For how much text fits, use the context section above — not this table.

ProviderGoogle Gemma 4 31b inGoogle Gemma 4 31b outKimi K2 Thinking Maas inKimi K2 Thinking Maas out
Aws Bedrock$0.140/M$0.400/M——
Google Vertex——$0.600/M$2.50/M

Frequently asked questions

Kimi K2 Thinking Maas has a larger context window: 256K tokens vs 256K. For long documents, large codebases, or extended agent sessions, the larger context window reduces the need to chunk inputs or summarize history.

Powered by Mem0

Use a smaller model.
Get better results.

Mem0 gives your AI long-term memory so you stop re-sending context on every call. That means you can use a smaller, faster, cheaper model — and still get better answers.

Calculate your savingsLearn about Mem0

Example: a multi-turn chat session

Without Mem0~128K tokens sent
Full history
Repeated info
Old context
With Mem0~20K tokens sent
Key memories
Current turn

80% less to send — works with any model