Google: Gemini 1.5 Flash-8B

Gemini 1.5 Flash-8B is a smaller and faster variant of Gemini 1.5 Flash. Google's developer blog announced it as production-ready on October 3, 2024 under the model id gemini-1.5-flash-8b, calling it "our latest Flash variant" and aiming it at simple, high-volume tasks. It is an early-generation model: Google's current model pages are built around the Gemini 3 family, and newer low-cost options such as Gemini 3.5 Flash-Lite now fill the same cheap, high-throughput niche.

Status

Google's Gemini API deprecations page currently lists only Gemini 2.0, 2.5 and 3 models, and it does not mention Gemini 1.5 Flash-8B. We could not confirm an official shutdown date for this model from a Google documentation page, so none is stated here. Before relying on gemini-1.5-flash-8b in any project, call the API to check that the model id still resolves, and prefer a current Flash-Lite model for new work.

Key features

According to Google's announcement post:

• Lower price: $0.0375 per 1M input tokens and $0.15 per 1M output tokens for prompts under 128K, with cached prompts at $0.01 per 1M. Google says this is a 50% lower price than Gemini 1.5 Flash.
• Higher rate limits: up to 4,000 requests per minute, double those of Gemini 1.5 Flash.
• Lower latency: Google says latency is lower than 1.5 Flash on small prompts.
• Task fit: the post says it performs especially well on chat, transcription and long-context translation, and nearly matches 1.5 Flash on many benchmarks.

Pricing above is as published at launch on October 3, 2024. It is not current pricing.

Use cases

• High-volume classification, tagging and extraction on simple inputs.
• Chat and transcription workloads where cost per request matters most.
• Long-context summarization and translation at low cost.

Getting started

1. Check that the gemini-1.5-flash-8b model id still works for your API key.
2. For new projects, start from the Gemini API models page. As of October 28, 2026, Gemini 3.5 Flash-Lite is described there as Google's most cost-effective 3.5 model.
3. Re-run your own evaluation before swapping models, since quality and output style differ between generations.

FAQ

What is Gemini 1.5 Flash-8B?

A smaller variant of Gemini 1.5 Flash, announced by Google on October 3, 2024 for simple, high-volume tasks.

How much did Gemini 1.5 Flash-8B cost?

At launch, $0.0375 per 1M input tokens and $0.15 per 1M output tokens for prompts under 128K, according to Google's announcement.

Is Gemini 1.5 Flash-8B still available?

Google's current deprecations page does not list it, and we could not confirm an official shutdown date. Test the model id directly, and plan to migrate.

What should I use instead?

Gemini 3.5 Flash-Lite is a current low-cost option, and Gemini 2.0 Flash is an earlier one.

Alternatives

• Gemini 3.5 Flash-Lite: a current low-cost Google model.
• Gemini 3.1 Flash-Lite: the previous Flash-Lite generation.
• Gemini 2.0 Flash: an earlier Flash model.

Conclusion

Gemini 1.5 Flash-8B was Google's cheapest Gemini 1.5 model at its launch. It is best treated as a historical reference, and new projects should choose from the current lineup on the Gemini API models page or in the models category.