Gemini 4 Argon and Gemini 3.8 Flash are available on BetterToken
Compare Gemini 4 Argon and Gemini 3.8 Flash, their official specifications, BetterToken APIs, cache limits, and current discount.
Updated: October 28, 2026
Gemini 4 Argon and Gemini 3.8 Flash are available on BetterToken. Use your existing BetterToken API Key with gemini-4-argon or gemini-3.8-flash.
Choose a model
| Model | Model ID | Suitable tasks |
|---|---|---|
| Gemini 4 Argon | gemini-4-argon | Complex software engineering, mathematical and scientific analysis, and multi-step reasoning. |
| Gemini 3.8 Flash | gemini-3.8-flash | Coding, content processing, and repeated high-volume tasks that balance speed and cost. |
High-volume tasks describe a use case, not availability of Google’s Batch API or batch discounts on BetterToken.
Official specifications and BetterToken support
Google lists text, image, video, audio, and PDF input with text output for both models. Thinking levels are low, medium (default), and high.
- Gemini 3.8 Flash: up to
1,048,576input tokens and65,536output tokens. - Gemini 4 Argon: input, thinking, and output share a
1,048,576-token window. Ordinary synchronous requests allow up to65,536output tokens.
These are Google specifications. BetterToken has not fully validated every modality or the complete context window. Google’s built-in tools, Batch/Flex/Priority, and Argon’s long decode are outside the BetterToken access paths validated here.
BetterToken provides native Gemini and OpenAI-compatible Chat Completions, supporting text, video, streaming, function calls, and tool-result submission. Responses does not support these models. Codex CLI and other Responses-only clients cannot call them.
Quick setup
- Use your existing BetterToken API Key, or register and create one.
- Choose a protocol and authentication field below. Set Model ID to
gemini-4-argonorgemini-3.8-flash.
| Protocol | Base URL | Authentication header |
|---|---|---|
| Native Gemini / Google GenAI SDK | https://bettertoken.ai | x-goog-api-key: YOUR_BETTERTOKEN_API_KEY |
| OpenAI-compatible Chat Completions | https://bettertoken.ai/v1 | Authorization: Bearer YOUR_BETTERTOKEN_API_KEY |
Native requests use POST /v1beta/models/{model}:generateContent; streaming uses POST /v1beta/models/{model}:streamGenerateContent?alt=sse. Do not append /v1beta to the SDK Base URL. Chat Completions uses POST /v1/chat/completions.
See the Gemini API guide for complete examples, video input, and tool-result submission.
Current pricing and caching
As of October 28, 2026, BetterToken input and output prices for both models are 32% below Google’s current introductory rates. Check the model plaza for live USD, RUB, and cache-read prices.
Google’s Argon introductory pricing runs through November 30, 2026; Flash introductory pricing runs through December 31, 2026. These are Google’s promotion dates. Future BetterToken prices remain subject to the model plaza.
Cache hits are not guaranteed. Argon officially supports implicit caching only, not explicit cache creation. BetterToken does not expose an explicit cache-creation API. Evaluate caching from your actual usage records.
