Gemini 3.8 Flash: Google's strongest reasoning and coding model at Flash speed and price

Google released Gemini 3.8 Flash on September 2, 2026. It is their best reasoning and coding model yet, and it runs at the same speed and cost as 3.7 Flash, which came out three weeks earlier. This is Google's third Flash release in six weeks.

The model comes in two variants. Gemini 3.8 Flash is the general-purpose workhorse. Gemini 3.8 Flash Cyber is a cybersecurity specialist restricted to trusted defenders through Google's Fairwind Program. Both share the same core intelligence, tuned through long-running agentic training loops that recursively evaluate and refine the underlying models. The coding and reasoning gains across this shared core come from several innovations, including rigorous training in the cybersecurity domain.

Performance gains over 3.7 Flash

3.8 Flash delivers substantial gains over 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning. On several benchmarks it approaches or matches the performance of larger, more expensive frontier models.

On DeepSWE v1.1, a benchmark for long-horizon software engineering, 3.8 Flash outperforms most larger frontier models in autonomously solving complex engineering problems end to end, at a fraction of the cost.

In professional domains the results are similar. 3.8 Flash outperforms 3.7 Flash and other frontier models on Vals Finance Agent V2 and Harvey's Legal Agent Benchmark, two benchmarks that test advanced analysis and reporting in quantitative and legal fields. It scores 54.9% on HLE-Verified, which measures multi-step reasoning across STEM, humanities, and professional fields.

How it works

The main design choice behind these gains: 3.8 Flash works harder. On complex tasks it runs extra reasoning steps and calls tools iteratively. At higher effort levels it may consume more tokens to maximize performance.

For workloads where compute efficiency matters more than peak quality, developers can use lower effort levels to reduce token overhead. Gemini 3.7 Flash also remains fully supported for efficiency-first workloads.

Pricing

The introductory price matches 3.7 Flash: $0.75 per million input tokens and $3.75 per million output tokens. This price expires on December 31, 2026. Starting January 1, 2027, the rate becomes $1.50 per million input tokens and $7.50 per million output tokens.

Safety

3.8 Flash ships with safeguards against misuse in chemical, biological, radiological, and nuclear domains, and against cyber offense, following Google's Frontier Safety Framework. Google also reports a significant improvement in prompt injection robustness as measured by Gray Swan, which protects users from prompt-injection attacks.

Try Gemini 3.8 Flash on vega.chat

Gemini 3.8 Flash is available now on vega.chat and through the api.vega.chat API. You can test it in the chat interface, compare its responses side by side with other models, or integrate it into your applications via the OpenRouter-compatible API at api.vega.chat.