Gemini 3.5 Flash-Lite vs Kimi K3: Specs & Benchmark Comparison

Gemini 3.5 Flash-Lite is developed by Google, while Kimi K3 comes from Moonshot AI. Both were released in July 2026. Kimi K3 has a published size of about 2.8 trillion parameters; Google has not disclosed the parameter count of Gemini 3.5 Flash-Lite.

The two models share 2 published benchmarks. Kimi K3 leads on 2 of them. The widest gaps are on Terminal-Bench 2.1, where Kimi K3 scores 88.3% against 54.0%; CharXiv-R, where Kimi K3 scores 91.3% against 76.5%. Averaged across everything we track, Gemini 3.5 Flash-Lite sits at 53.2% and Kimi K3 at 67.8%.

Gemini 3.5 Flash-Lite is the cheaper API at $0.30 per million input tokens and $2.50 per million output tokens, roughly 10 times cheaper than Kimi K3 at $3 and $15. Both accept a context window of about 1M tokens. Both accept text and images as input.

CharacteristicGemini 3.5 Flash-LiteKimi K3
CompanyGoogleMoonshot AI
Release DateJuly 21, 2026July 16, 2026
Parameters2.8T
MultimodalYesYes
Context (input)1.0M1.0M
Context (output)66K1.0M
Input Price / 1M$0.30$3.00
Output Price / 1M$2.50$15.00
Average Score53.2%67.8%
Benchmarks
Terminal-Bench 2.154.0%88.3%
CharXiv-R76.5%91.3%

Visual Benchmark Comparison

Gemini 3.5 Flash-Lite
Kimi K3
Terminal-Bench 2.10.5 vs 0.9
0.5
0.9
CharXiv-R0.8 vs 0.9
0.8
0.9

Verdict

Gemini 3.5 Flash-Lite leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: Gemini 3.5 Flash-Lite — 0.5, Kimi K3 — 0.7.

API Cost

Gemini 3.5 Flash-Lite is 6.4x cheaper: input $0.30/1M vs $3.00/1M tokens.

Context Window

Gemini 3.5 Flash-Lite supports a larger context: 1M vs 1M tokens.

Recency

Both models were released around the same time: 7/21/2026 and 7/16/2026.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — Gemini 3.5 Flash-Lite or Kimi K3?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — Gemini 3.5 Flash-Lite or Kimi K3?
Gemini 3.5 Flash-Lite is cheaper for input: $0.30 per 1M tokens vs $3.00.
Which has a larger context window — Gemini 3.5 Flash-Lite or Kimi K3?
Gemini 3.5 Flash-Lite supports a larger context: 1,048,576 tokens vs 1,000,000.

The Gemini 3.5 Flash-Lite and Kimi K3 comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Gemini 3.5 Flash-Lite or Kimi K3 page. See also the complete list of AI model comparisons.