Model comparison
Kimi K3 vs Gemma 4
- Input
- $3.00
- Output
- $15.00
- Context
- 1M
- Released
- 2026-07-16
- Input
- $0.12 / 1M tokens
- Output
- $0.35 / 1M tokens
- Context
- 256K
- Released
- 2026-04-02
Kimi K3 scores higher on 2 of the 2 benchmarks both models report; Gemma 4 is cheaper on input tokens ($0.1 vs $3 per 1M, 25×); Kimi K3 has the larger context window (1M vs 256K).
Specs and pricing side by side
| Spec | Kimi K3 | Gemma 4 |
|---|---|---|
| Maker | Moonshot AI (Kimi) | |
| Released | 2026-07-16 | 2026-04-02 |
| Context window | 1M | 256K |
| Input price | $3.00 | $0.12 / 1M tokens |
| Output price | $15.00 | $0.35 / 1M tokens |
| Parameters | 2.8T total · ~16 of 896 experts active per token | 30.7B (31B dense flagship); family spans E2B/E4B on-device and 26B MoE (3.8B active) |
| License | Kimi K3 License (custom, weights published on Hugging Face 2026-07-26) | Apache 2.0 |
| Weights | Open weights | Open weightssame |
| Modalities | Text, Vision, Video | Text, Vision, Audio, Video |
| Knowledge cutoff | not published | January 2025 |
Not published by either maker: max output.
Benchmarks both models report
| Benchmark | Kimi K3 | Gemma 4 |
|---|---|---|
| GPQA-Diamond | 93.5% | 84.3% |
| MMMU-Pro | 81.6% | 76.9% |
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 GPQA-Diamond, MMMU-Pro (it scores higher on 2 shared benchmarks).
- You need the longer context (1M vs 256K).
- Your use case is: reach for it for long-horizon agentic coding and browser/tool workflows that need a very long context.
Pick Gemma 4 if…
- Cost matters: input tokens cost 25× less ($0.1 vs $3 per 1M).
- You need Audio input, which Kimi K3 lacks.
- Your use case is: self-hosted and on-premise LLM deployments where data must stay private.
FAQ
Which is better, Kimi K3 or Gemma 4?
Kimi K3 scores higher on 2 of the 2 benchmarks both models report; Gemma 4 is cheaper on input tokens ($0.1 vs $3 per 1M, 25×); Kimi K3 has the larger context window (1M vs 256K). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is Gemma 4?
Kimi K3: $3.00 input, $15.00 output. Gemma 4: $0.12 / 1M tokens input, $0.35 / 1M tokens output (list API prices as published by each maker).
What are the context windows of Kimi K3 and Gemma 4?
Kimi K3 offers 1M; Gemma 4 offers 256K.
Is Kimi K3 newer than Gemma 4?
Kimi K3 was released 2026-07-16; Gemma 4 was released 2026-04-02.