Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Floor (cheapest).
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | Gemini 3.8 Flash (High) Intelligence Index | 30.8 |
| Artificial Analysis | Gemini 3.8 Flash (High) Coding Index | 76.3 |
| Artificial Analysis | Gemini 3.8 Flash (High) Agentic Index | 40.2 |
| Artificial Analysis | Gemini 3.8 Flash (High) GPQA Diamond | 95.3% |
| Artificial Analysis | Gemini 3.8 Flash (High) HLE | 47.8% |
| Artificial Analysis | Gemini 3.8 Flash (High) AA-LCR | 81.3% |
| Artificial Analysis | Gemini 3.8 Flash (High) τ-Bench Banking | 44.9% |
| Artificial Analysis | Gemini 3.8 Flash (High) GDPval-AA | 47.3% |
| Artificial Analysis | Gemini 3.8 Flash (High) CritPt | 18.3% |
| Artificial Analysis | Gemini 3.8 Flash (High) SciCode | 56.6% |
| Artificial Analysis | Gemini 3.8 Flash (High) Terminal-Bench 4.0 | 19.7% |
| Artificial Analysis | Gemini 3.8 Flash (High) Terminal-Bench 2.1 | 87.6% |
| Artificial Analysis | Gemini 3.8 Flash (High) AA-Omniscience Accuracy | 50.5% |
| Artificial Analysis | Gemini 3.8 Flash (High) AA-Omniscience Non-Hallucination Rate | 44.9% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) Intelligence Index | 30.1 |
| Artificial Analysis | Gemini 3.8 Flash (Medium) Coding Index | 74.1 |
| Artificial Analysis | Gemini 3.8 Flash (Medium) Agentic Index | 39.7 |
| Artificial Analysis | Gemini 3.8 Flash (Medium) GPQA Diamond | 93.5% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) HLE | 42.1% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) AA-LCR | 84.0% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) τ-Bench Banking | 45.8% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) GDPval-AA | 46.6% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) CritPt | 12.3% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) SciCode | 55.1% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) Terminal-Bench 4.0 | 19.7% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) Terminal-Bench 2.1 | 83.9% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) AA-Omniscience Accuracy | 49.6% |
| Artificial Analysis | Gemini 3.8 Flash (Medium) AA-Omniscience Non-Hallucination Rate | 48.5% |
| Artificial Analysis | Gemini 3.8 Flash (Low) Intelligence Index | 22.5 |
| Artificial Analysis | Gemini 3.8 Flash (Low) Coding Index | 73.5 |
| Artificial Analysis | Gemini 3.8 Flash (Low) Agentic Index | 32.6 |
| Artificial Analysis | Gemini 3.8 Flash (Low) GPQA Diamond | 92.0% |
| Artificial Analysis | Gemini 3.8 Flash (Low) HLE | 37.1% |
| Artificial Analysis | Gemini 3.8 Flash (Low) AA-LCR | 80.7% |
| Artificial Analysis | Gemini 3.8 Flash (Low) τ-Bench Banking | 33.2% |
| Artificial Analysis | Gemini 3.8 Flash (Low) GDPval-AA | 40.8% |
| Artificial Analysis | Gemini 3.8 Flash (Low) CritPt | 4.0% |
| Artificial Analysis | Gemini 3.8 Flash (Low) SciCode | 55.0% |
| Artificial Analysis | Gemini 3.8 Flash (Low) Terminal-Bench 4.0 | 10.1% |
| Artificial Analysis | Gemini 3.8 Flash (Low) Terminal-Bench 2.1 | 83.1% |
| Artificial Analysis | Gemini 3.8 Flash (Low) AA-Omniscience Accuracy | 48.3% |
| Artificial Analysis | Gemini 3.8 Flash (Low) AA-Omniscience Non-Hallucination Rate | 31.9% |
| Design Arena | Gemini 3.8 Flash Agents Arena Androidnative Elo | 1092 |
| Design Arena | Gemini 3.8 Flash Agents Arena Blender3d Elo | 1198 |
| Design Arena | Gemini 3.8 Flash Agents Arena Codetovideo Elo | 1236 |
| Design Arena | Gemini 3.8 Flash Agents Arena Full Stack Elo | 1244 |
| Design Arena | Gemini 3.8 Flash Agents Arena Godotgamedev Elo | 1273 |
| Design Arena | Gemini 3.8 Flash Agents Arena Mobile Apps Elo | 1232 |
| Design Arena | Gemini 3.8 Flash Agents Arena Pdf Elo | 1220 |
| Design Arena | Gemini 3.8 Flash Agents Arena Python-Pptxslides Elo | 1138 |
| Design Arena | Gemini 3.8 Flash Agents Arena Spreadsheet Elo | 1070 |
| Design Arena | Gemini 3.8 Flash Agents Arena Webapps Elo | 1226 |
| Design Arena | Gemini 3.8 Flash Models Arena 3D Elo | 1291 |
| Design Arena | Gemini 3.8 Flash Models Arena Asciiart Elo | 1269 |
| Design Arena | Gemini 3.8 Flash Models Arena Code Categories Elo | 1302 |
| Design Arena | Gemini 3.8 Flash Models Arena Data Visualization Elo | 1273 |
| Design Arena | Gemini 3.8 Flash Models Arena Game Development Elo | 1305 |
| Design Arena | Gemini 3.8 Flash Models Arena SVG Elo | 1293 |
| Design Arena | Gemini 3.8 Flash Models Arena UI Component Elo | 1313 |
| Design Arena | Gemini 3.8 Flash Models Arena Website Elo | 1299 |
| Vals AI | Gemini 3.8 Flash BioMysteryBench | 62.2% |
| Vals AI | Gemini 3.8 Flash Code Migration v1.1 | 38.6% |
| Vals AI | Gemini 3.8 Flash CyberBench v1.1 | 43.8% |
| Vals AI | Gemini 3.8 Flash EMB v1.1 | 72.9% |
| Vals AI | Gemini 3.8 Flash Finance Agent v2.1 | 65.8% |
| Vals AI | Gemini 3.8 Flash Vals Finance Index | 60.5% |
| Vals AI | Gemini 3.8 Flash GPQA Diamond | 94.4% |
| Vals AI | Gemini 3.8 Flash HLAB-Vals-1.1 | 12.1% |
| Vals AI | Gemini 3.8 Flash Immigration | 63.4% |
| Vals AI | Gemini 3.8 Flash IOI | 56.9% |
| Vals AI | Gemini 3.8 Flash LiveCodeBench | 89.5% |
| Vals AI | Gemini 3.8 Flash LegalBench | 87.0% |
| Vals AI | Gemini 3.8 Flash Vals Legal Index | 69.8% |
| Vals AI | Gemini 3.8 Flash Legal Research Bench | 38.9% |
| Vals AI | Gemini 3.8 Flash MedCode | 48.1% |
| Vals AI | Gemini 3.8 Flash MedScribe | 84.5% |
| Vals AI | Gemini 3.8 Flash MMLU Pro | 90.2% |
| Vals AI | Gemini 3.8 Flash MMMU Pro | 89.1% |
| Vals AI | Gemini 3.8 Flash MortgageTax | 65.3% |
| Vals AI | Gemini 3.8 Flash MysteryMechanism | 36.5% |
| Vals AI | Gemini 3.8 Flash ProgramBench | 1.0% |
| Vals AI | Gemini 3.8 Flash ProofBench v1.1 | 48.0% |
| Vals AI | Gemini 3.8 Flash Public Benefits Bench v1.1 | 65.3% |
| Vals AI | Gemini 3.8 Flash Vals RSI Index | 19.9% |
| Vals AI | Gemini 3.8 Flash SAGE | 88.2% |
| Vals AI | Gemini 3.8 Flash Vals Science Index | 35.8% |
| Vals AI | Gemini 3.8 Flash SkillsBench | 58.0% |
| Vals AI | Gemini 3.8 Flash SRE Bench | 11.1% |
| Vals AI | Gemini 3.8 Flash SWE-bench | 80.0% |
| Vals AI | Gemini 3.8 Flash SWE-bench Multimodal | 40.8% |
| Vals AI | Gemini 3.8 Flash Tax Agent Bench | 32.1% |
| Vals AI | Gemini 3.8 Flash TaxEval v2 | 74.4% |
| Vals AI | Gemini 3.8 Flash Terminal-Bench 4.0 | 19.7% |
| Vals AI | Gemini 3.8 Flash Terminal-Bench-Science | 8.6% |
| Vals AI | Gemini 3.8 Flash Vals Index | 56.0% |
| Vals AI | Gemini 3.8 Flash Vibe Code Bench 1-100 | 18.8% |
| Vals AI | Gemini 3.8 Flash Vibe Code Bench v1.1 | 78.7% |
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
Token volume and request traffic to this model over time.
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
Not currently available through EU in-region routing. In-region routing docs
Guides, cookbooks, blog posts, and tools for building with Google: Gemini 3.8 Flash (batch) on OpenRouter.
50% off | $0.75$0.375 | $3.75$1.875 | $0.075$0.0375 | $0.075$0.0375 | -- | -- |
|---|
Not enough uptime data to display yet.
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.
Not enough data for apps using this model yet.
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
The Gemini 3.8 Flash (batch) documentation on OpenRouter includes the How to Extract Tables from a PDF with an LLM blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.
Gemini 3.8 Flash (batch) costs $0.375/M input tokens and $1.875/M output tokens, with separate rates for Cache Read at $0.0375/M tokens, Cache Write at $0.04167/M tokens, Image Input at $0.375/M tokens, Input Audio at $0.375/M tokens, Input Audio Cache at $0.0375/M tokens, and Web Search at $14.00/1K calls.
Gemini 3.8 Flash (batch) has a 1,048,576 token context window. It supports up to 65,536 completion tokens.
Yes. Gemini 3.8 Flash (batch) accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
Gemini 3.8 Flash (batch) accepts text, images, video, files such as PDFs, and audio as input and returns text.
Gemini 4 Argon, Nano Banana 2.1, Gemini 3.7 Flash and 26 more are other text models from Google.
Gemini 3.8 Flash (batch) was released on September 2, 2026.
