The AI arena is free today

Open Superagent

Gemini 3.8 Flash vs Kimi K2.5

Gemini 3.8 Flash leads the LLM Stats Score 51.0 to 39.1. Kimi K2.5 is 1.3x cheaper per token.

Google · Moonshot AI · Updated for 2026

Which is better?

Gemini 3.8 Flash leads the overall LLM Stats Score 51.0 to 39.1, ranking #15 overall.

In the 2 individual benchmarks reported for both models, Gemini 3.8 Flash wins 2; this is a narrower head-to-head signal than the composite indexes.

On price, Kimi K2.5 is roughly 1.3x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Gemini 3.8 Flash also accepts a larger context window (1,048,576 input tokens), making it the stronger choice for long documents and large codebases.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose Gemini 3.8 Flash

  • overall performance matters — it scores 51.0 and ranks #15 on LLM Stats
  • your work emphasizes reasoning and coding — it leads those capability indexes
  • you value its reported benchmark strengths — it wins 2 of 2 exact shared results
  • you process long inputs — it offers a 1,048,576 token context window
  • you want the most recent training data — it shipped Sep 2026

Choose Kimi K2.5

  • cost matters — it's about 1.3x cheaper per token
  • you need open weights you can self-host or fine-tune

At a glance

The differences that matter most.

Gemini 3.8 Flash
Google
Kimi K2.5
Moonshot AI
Core performance indexes
LLM Stats Score
51.0
#15
39.1
#58
Reasoning
46.9
#24
39.1
#55
Coding
39.3
#14
25.2
#65
Agents
35.8
#16
17.8
#65
Cost, coverage & limits
Benchmark wins
2 of 2
0 of 2
Input price
$0.75 / M
$0.60 / M
Output price
$3.75 / M
$3.00 / M
Context window
1,048,576
262,100

Capability indexes

Additional strengths measured across groups of related public benchmarks

2 shared
Index
Gemini 3.8 Flash
Kimi K2.5
Vision
33.8#15
27.6#35
Multimodal
34.8#12
29.3#29
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

13 reported for Gemini 3.8 Flash · 40 for Kimi K2.5

2 shared

Gemini 3.8 Flash outperforms in 2 benchmarks (CharXiv-R, LVBench), while Kimi K2.5 is better at 0 benchmarks.

Gemini 3.8 Flash significantly outperforms across most benchmarks.

Fri Sep 04 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Kimi K2.5 costs less

For input processing, Gemini 3.8 Flash ($0.75/1M tokens) is 1.3x more expensive than Kimi K2.5 ($0.60/1M tokens).

For output processing, Gemini 3.8 Flash ($3.75/1M tokens) is 1.3x more expensive than Kimi K2.5 ($3.00/1M tokens).

In conclusion, Gemini 3.8 Flash is more expensive than Kimi K2.5.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Fri Sep 04 2026 • llm-stats.com
Google
Gemini 3.8 Flash
Input tokens$0.75
Output tokens$3.75
Best providerGoogle
Moonshot AI
Kimi K2.5
Input tokens$0.60
Output tokens$3.00
Best providerFireworks
Notice missing or incorrect data?Start an Issue→

Context Window

Maximum input and output token capacity

Gemini 3.8 Flash accepts 1,048,576 input tokens compared to Kimi K2.5's 262,100 tokens. Kimi K2.5 can generate longer responses up to 262,100 tokens, while Gemini 3.8 Flash is limited to 65,536 tokens.

Google
Gemini 3.8 Flash
Input1,048,576 tokens
Output65,536 tokens
Moonshot AI
Kimi K2.5
Input262,100 tokens
Output262,100 tokens
Fri Sep 04 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Both Gemini 3.8 Flash and Kimi K2.5 support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Gemini 3.8 Flash

Text
Images
Audio
Video

Kimi K2.5

Text
Images
Audio
Video

License

Usage and distribution terms

Gemini 3.8 Flash is licensed under a proprietary license, while Kimi K2.5 uses MIT.

License differences may affect how you can use these models in commercial or open-source projects.

Gemini 3.8 Flash

Proprietary

Closed source

Kimi K2.5

MIT

Open weights

Release Timeline

When each model was launched

Gemini 3.8 Flash was released on 2026-09-02, while Kimi K2.5 was released on 2026-01-27.

Gemini 3.8 Flash is 7 months newer than Kimi K2.5.

Gemini 3.8 Flash

Sep 2, 2026

2 days ago

7mo newer
Kimi K2.5

Jan 27, 2026

7 months ago

Knowledge Cutoff

When training data ends

Gemini 3.8 Flash has a documented knowledge cutoff of 2026-03-31, while Kimi K2.5's cutoff date is not specified.

We can confirm Gemini 3.8 Flash's training data extends to 2026-03-31, but cannot make a direct comparison without Kimi K2.5's cutoff date.

Gemini 3.8 Flash

Mar 2026

Kimi K2.5

—

Provider Availability

Gemini 3.8 Flash is available from Google. Kimi K2.5 is available from Fireworks, Moonshot AI.

Gemini 3.8 Flash

google logo
Google
Input Price:Input: $0.75/1MOutput Price:Output: $3.75/1M

Kimi K2.5

fireworks logo
Fireworks
Input Price:Input: $0.60/1MOutput Price:Output: $3.00/1M
moonshot logo
Unknown Organization
Input Price:Input: $0.60/1MOutput Price:Output: $3.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Gemini 3.8 Flash and Kimi K2.5 side-by-side, then vote on the output you prefer.

Gemini 3.8 Flash
✓ Preferred
Kimi K2.5
Open in Playground

FAQ

Common questions about Gemini 3.8 Flash vs Kimi K2.5.

Which is better, Gemini 3.8 Flash or Kimi K2.5?

Gemini 3.8 Flash leads the LLM Stats Score 51.0 to 39.1. Gemini 3.8 Flash is made by Google and Kimi K2.5 is made by Moonshot AI. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Gemini 3.8 Flash compare to Kimi K2.5 in benchmarks?

Gemini 3.8 Flash scores Terminal-Bench 2.1: 89.4%, LVBench: 87.1%, CharXiv-R: 86.2%, LABBench2: 86.2%, DeepSWE 1.1: 73.7%. Kimi K2.5 scores AIME 2025: 96.1%, HMMT 2025: 95.4%, InfoVQAtest: 92.6%, OCRBench: 92.3%, MathVista-Mini: 90.1%.

Is Gemini 3.8 Flash cheaper than Kimi K2.5?

Kimi K2.5 is 1.3x cheaper for input tokens. Gemini 3.8 Flash costs $0.75/M input and $3.75/M output via google. Kimi K2.5 costs $0.60/M input and $3.00/M output via fireworks.

What are the context window sizes for Gemini 3.8 Flash and Kimi K2.5?

Gemini 3.8 Flash supports 1.0M tokens and Kimi K2.5 supports 262K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Gemini 3.8 Flash and Kimi K2.5?

Key differences include LLM Stats Score (51.0 vs 39.1), context window (1.0M vs 262K), input pricing ($0.75 vs $0.60/M), licensing (Proprietary vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Gemini 3.8 Flash and Kimi K2.5?

Gemini 3.8 Flash is developed by Google and Kimi K2.5 is developed by Moonshot AI.

Related comparisons