Google model
Gemini 3.8 Flash API
google/gemini-3.8-flash is the lowest-priced Flash id of Google's current 3.x generation in the NextModel catalog: $0.75 per million input tokens, $3.75 per million output tokens, $0.08 per million cache hits, and a 1,000,000 token context window. That is exactly half the published row of gemini-3.6-flash and gemini-3.7-flash ($1.50 / $7.50 / $0.15). The trade is integration: the catalog lists this id on the gemini API surface only, so OpenAI SDK chat-completions code must either move to the Gemini request shape or pick a different id — google/gemini-3.6-flash takes both surfaces at the higher rate. Capabilities listed are Thinking, Function Calling, and Vision. No first-party benchmark scores or latency band are published for this id.
ThinkingFunction CallingVision
What is Gemini 3.8 Flash API in NextModel?
google/gemini-3.8-flash is the lowest-priced Flash id of Google's current 3.x generation in the NextModel catalog: $0.75 per million input tokens, $3.75 per million output tokens, $0.08 per million cache hits, and a 1,000,000 token context window. That is exactly half the published row of gemini-3.6-flash and gemini-3.7-flash ($1.50 / $7.50 / $0.15). The trade is integration: the catalog lists this id on the gemini API surface only, so OpenAI SDK chat-completions code must either move to the Gemini request shape or pick a different id — google/gemini-3.6-flash takes both surfaces at the higher rate. Capabilities listed are Thinking, Function Calling, and Vision. No first-party benchmark scores or latency band are published for this id.
Gemini 3.8 Flash via NextModel: 1M-token context; $0.75 / 1M tokens in / $3.75 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.
Best use cases
- Gemini-shape codebases that want current-generation Flash pricing at $0.75 / $3.75 per million tokens — half of what 3.6 and 3.7 publish.
- Repeated-context jobs — long instructions, recurring document sets — that can earn the listed $0.08 per million cache-hit rate, half of the 3.6/3.7 cache row.
- Vision or tool-call workloads that need a Thinking entry below a $1.00 input rate: this row lists Thinking, Function Calling, and Vision, while gemini-3.1-flash-lite's row lists only Vision and Function Calling at $0.25 / $1.50.
When not to use
- OpenAI SDK chat-completions code you will not adapt: this id is gemini-only, and the mismatch surfaces at request time, not at signup. Use google/gemini-3.6-flash if the both-surface row matters more than the price.
- Absolute price-floor traffic: gemini-3.1-flash-lite lists $0.25 / $1.50 per million tokens, and deepseek/deepseek-v4-flash lists $0.14 / $0.28.
- Pipelines that require Streaming or Structured Outputs entries on the catalog row: this id lists Thinking, Function Calling, and Vision only.
OpenAI-compatible code example
request.sh
curl https://api.nextmodel.app/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $NEXTMODEL_API_KEY" \
-d '{"model":"google/gemini-3.8-flash","messages":[{"role":"user","content":"Reply with OK."}]}'Switching models only changes the model field.
Similar alternatives
AnthropicCatalog
Claude Fable 5.1 via NextModel: 1M-token context; $10 / 1M tokens in / $50 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
OpenAICatalog
GPT-6 AstraNew
GPT-6 Astra via NextModel: 1.1M-token context; Starting at $10 / 1M tokens in / $50 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
QwenCatalog
Qwen3.8-Flash via NextModel: 1M-token context; $0.12 / 1M tokens in / $0.4 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
Reading
Articles that explain Gemini 3.8 Flash
LLM Gateway: Compare Options and Alternatives
Gemini-shape and OpenAI-shape ids resolve through one base URL and key.
Zhao Yu'an
How to estimate AI API cost before you ship
Half-price does not mean half-bill — estimate from your own traffic.
Lin Qiming
FAQ
Gemini 3.8 Flash API questions
What model id and endpoint do I send for Gemini 3.8 Flash?
Send google/gemini-3.8-flash to the Gemini surface under https://api.nextmodel.app/v1/gemini. The catalog lists this id on the gemini API only — it is not on the OpenAI-compatible chat completions endpoint. OpenAI SDK code should either use the Gemini request shape or switch to google/gemini-3.6-flash, which serves both surfaces.
What does the catalog list for price and context?
$0.75 per million input tokens, $3.75 per million output tokens, and $0.08 per million cache hits. Context window: 1,000,000 tokens. Every line is half of gemini-3.6-flash and gemini-3.7-flash's published row.
What should I treat as unverified?
Latency band and first-party benchmark scores are unpublished for this id. The Thinking, Function Calling, and Vision entries come from the catalog row, and the catalog publishes no quality scores behind them.
Start now
Ready to start building?
Passwordless signup, pay-as-you-go billing, and one OpenAI-compatible endpoint for every model. Create your key and send your first request in minutes.