// head_to_head

Gemini 3.1 Pro Preview vs Kimi K3

Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.

OpenRouter + LiveBenchAll comparisonsFull leaderboard

Kimi K3 scores higher, Gemini 3.1 Pro Preview costs less — it depends on your workload.

Kimi K3 is ahead by 2.2 points overall, and Gemini 3.1 Pro Preview lists 33% cheaper per blended million tokens. Whether 2.2 points is worth that depends on how much a wrong answer costs you. Gemini 3.1 Pro Preview also leads on measured cost per point of capability, at $0.1567 per point.

google

Gemini 3.1 Pro Preview

Blended / 1M
$4.50
Context
1.0M
Released
Feb 19, 2026
Overall score
77.0
reasoningtool callingaudio inputfile inputimage inputvideo inputprompt caching

moonshotai

Kimi K3

Blended / 1M
$6.00
Context
1.0M
Released
Jul 16, 2026
Overall score
79.2
reasoningtool callingimage inputvideo inputprompt caching

Specs and pricing

MetricGemini 3.1 Pro PreviewKimi K3
LiveBench overall

Mean of the seven LiveBench category scores, 0–100. Higher is better.

77.079.2win
Cost per point

Measured benchmark spend divided by overall score — dollars per point of capability.

$0.1567win$0.1909
Blended price / 1M

3:1 input:output mix, the usual shape of production traffic.

$4.50win$6.00
Input price / 1M$2.00win$3.00
Output price / 1M$12.00win$15.00
Cached input / 1M

Price of an input token served from the prompt cache, where the provider publishes one.

$0.200win$0.300
Context window1.0M1.0M
Max output tokens66K

Benchmarks by category

An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Gemini 3.1 Pro Preview on top, Kimi K3 below, both out of 100.

Agentic coding
44.1
62.2
Coding
76.5
81.4
Reasoning
84.0
90.7
Mathematics
91.0
84.4
Data analysistoo close to call
78.5
78.7
Languagetoo close to call
85.4
85.5
Instruction following
79.1
71.4

What each one costs to run

Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.

WorkloadGemini 3.1 Pro PreviewKimi K3
Support chatbot

1.2K in / 400 out × 200K requests

$1310.40/mo$1725.60/mo
RAG assistant

8K in / 600 out × 100K requests

$1600.00/mo$2220.00/mo
Coding agent

40K in / 4K out × 20K requests

$1552.00/mo$2088.00/mo
Document extraction

20K in / 1.5K out × 50K requests

$2810.00/mo$3990.00/mo
Bulk classification

500 in / 20 out × 5M requests

$5300.00/mo$7650.00/mo
Run these two through the cost calculator

Which should you pick?

You are running this at volume

Gemini 3.1 Pro Preview

Lowest measured cost per point of capability at $0.1567 per point — the gap compounds with every request.

Quality matters more than the bill

Kimi K3

Highest overall LiveBench score of the two at 79.2.

The workload is coding or agentic work

Kimi K3

Leads on agentic coding — 62.2 against 44.1.

Gemini 3.1 Pro Preview vs Kimi K3 FAQ

Which is better, Gemini 3.1 Pro Preview or Kimi K3?

Kimi K3 scores higher, Gemini 3.1 Pro Preview costs less — it depends on your workload. Kimi K3 is ahead by 2.2 points overall, and Gemini 3.1 Pro Preview lists 33% cheaper per blended million tokens. Whether 2.2 points is worth that depends on how much a wrong answer costs you. Gemini 3.1 Pro Preview also leads on measured cost per point of capability, at $0.1567 per point.

Is Gemini 3.1 Pro Preview cheaper than Kimi K3?

Gemini 3.1 Pro Preview is cheaper. On a 3:1 input:output blend, Gemini 3.1 Pro Preview lists at $4.50 per million tokens and Kimi K3 at $6.00 — Gemini 3.1 Pro Preview is 33% cheaper. Input and output are priced separately — Gemini 3.1 Pro Preview charges $2.00 in and $12.00 out, Kimi K3 charges $3.00 and $15.00 — so the model that looks cheaper flips depending on how output-heavy your workload is.

Gemini 3.1 Pro Preview vs Kimi K3: which scores higher on benchmarks?

Gemini 3.1 Pro Preview scores 77.0 and Kimi K3 scores 79.2 overall on LiveBench, the mean of its seven categories. That is a 2.2-point lead for Kimi K3. Category scores differ from the overall figure — a model can lead on reasoning and trail on coding, which the per-category table above breaks out.

Which gives better value for money, Gemini 3.1 Pro Preview or Kimi K3?

Gemini 3.1 Pro Preview. Cost per point divides the measured dollars LiveBench spent running the benchmark by the score it earned, so it captures something token pricing misses: a reasoning model can emit many times more tokens than its per-token rate suggests. Gemini 3.1 Pro Preview works out at $0.1567 per point and Kimi K3 at $0.1909.

Does Gemini 3.1 Pro Preview or Kimi K3 have a bigger context window?

They are effectively the same — 1.0M for Gemini 3.1 Pro Preview and 1.0M for Kimi K3.

Do Gemini 3.1 Pro Preview and Kimi K3 support prompt caching?

Both publish a cached-input rate: $0.200 per million for Gemini 3.1 Pro Preview and $0.300 for Kimi K3, against full input rates of $2.00 and $3.00. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.

Related comparisons

How these numbers are produced

  • Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
  • ScoresLiveBench release 2026-06-25, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
  • Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
  • “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.

Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.