$ diff gemini-3-5-flash-lite minimax-m3
Gemini 3.5 Flash-Lite vs MiniMax M3
Google DeepMind and MiniMax, side by side. Every value below is the one recorded on the model’s own wiki page, with its citation — nothing is estimated.
Google DeepMindGemini 3.5 Flash-Lite
- context
- 1.0M
- weights
- closed
- $/M in
- $0.3
- $/M out
- $2.5
- context
- 1M
- weights
- open
- $/M in
- —
- $/M out
- —
What actually differs
- Context window
- Gemini 3.5 Flash-Lite takes 1.0M against 1M — modestly more room in a single request.
- Weights
- MiniMax M3 publishes weights (Open-weight (HuggingFace)); Gemini 3.5 Flash-Lite is API-only. That decides self-hosting, air-gapped deployment and fine-tuning before any capability question does.
- Recency
- Gemini 3.5 Flash-Lite shipped 50 days after MiniMax M3 (2026-07-21 vs 2026-06-01).
Full spec
| Attribute | Gemini 3.5 Flash-Lite | MiniMax M3 |
|---|---|---|
| Developer | Google DeepMind | MiniMax |
| Released | 2026-07-21 | 2026-06-01 |
| Context window | 1,048,576 tokens (1M) | 1,000,000 tokens (1M); min. 512K guaranteed |
| Pricing | $0.30/M input · $2.50/M output | Available via API; pricing on OpenRouter |
| License | proprietary | Open-weight (HuggingFace) |
| Availability | Gemini API, Google AI Studio, Vertex AI | API (global, including English) + open-weight on HuggingFace |
Values come from Gemini 3.5 Flash-Lite and MiniMax M3, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing. How these pages are produced.