Model comparison
Kimi K3 vs Gemini 3.5 Flash
- Input
- $3.00
- Output
- $15.00
- Context
- 1M
- Released
- 2026-07-16
- Input
- $1.50 / 1M tokens
- Output
- $9.00 / 1M tokens
- Context
- 1M
- Released
- 2026-05-19
Kimi K3 and Gemini 3.5 Flash split the 2 benchmarks both report evenly; Gemini 3.5 Flash is cheaper on input tokens ($1.5 vs $3 per 1M, 2×); Kimi K3 ships open weights while Gemini 3.5 Flash is API-only.
Specs and pricing side by side
| Spec | Kimi K3 | Gemini 3.5 Flash |
|---|---|---|
| Maker | Moonshot AI (Kimi) | |
| Released | 2026-07-16 | 2026-05-19 |
| Context window | 1M | 1Msame |
| Max output | not published | 64K |
| Input price | $3.00 | $1.50 / 1M tokens |
| Output price | $15.00 | $9.00 / 1M tokens |
| Parameters | 2.8T total · ~16 of 896 experts active per token | Not disclosed |
| License | Kimi K3 License (custom, weights published on Hugging Face 2026-07-26) | Proprietary |
| Weights | Open weights | API only |
| Modalities | Text, Vision, Video | Text, Vision, Audio, Video, PDF |
| Knowledge cutoff | not published | January 2025 |
Benchmarks both models report
| Benchmark | Kimi K3 | Gemini 3.5 Flash |
|---|---|---|
| Terminal Bench 2.1 | 88.3% | 76.2% |
| MMMU-Pro | 81.6% | 83.6% |
Scores are as published by each maker or its cited source; harness, reasoning effort and date can differ between rows, so treat small gaps as ties. Sources are listed on each model page.
Which one should you pick?
Pick Kimi K3 if…
- You care about Terminal Bench 2.1 (it scores higher on 1 shared benchmark).
- You want to self-host or fine-tune (open weights, Kimi K3 License (custom, weights published on Hugging Face 2026-07-26)).
- Your use case is: reach for it for long-horizon agentic coding and browser/tool workflows that need a very long context.
Pick Gemini 3.5 Flash if…
- You care about MMMU-Pro (it scores higher on 1 shared benchmark).
- Cost matters: input tokens cost 2× less ($1.5 vs $3 per 1M).
- You need Audio / PDF input, which Kimi K3 lacks.
- Your use case is: coding agents and autonomous multi-step software tasks (e.g. terminal and SWE-style workflows).
FAQ
Which is better, Kimi K3 or Gemini 3.5 Flash?
Kimi K3 and Gemini 3.5 Flash split the 2 benchmarks both report evenly; Gemini 3.5 Flash is cheaper on input tokens ($1.5 vs $3 per 1M, 2×); Kimi K3 ships open weights while Gemini 3.5 Flash is API-only. The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is Gemini 3.5 Flash?
Kimi K3: $3.00 input, $15.00 output. Gemini 3.5 Flash: $1.50 / 1M tokens input, $9.00 / 1M tokens output (list API prices as published by each maker).
What are the context windows of Kimi K3 and Gemini 3.5 Flash?
Kimi K3 offers 1M; Gemini 3.5 Flash offers 1M with up to 64K output.
Is Kimi K3 newer than Gemini 3.5 Flash?
Kimi K3 was released 2026-07-16; Gemini 3.5 Flash was released 2026-05-19.