OpenCode에서 Gemini 2.5 Flash-Lite 발견
확인됨근거 링크 10개
영어 원문

USD / 토큰 100만 개
| 입력 | 출력 | 캐시 입력 |
|---|---|---|
|
|
|
| 모델 | 근거 링크 | 입력/출력 가격 |
|---|---|---|
google/gemini-2.5-flash-litestandard | openrouter | $0.10 / $0.40 |
google/gemini-2.5-flash-litestandard | vercel-gateway | $0.10 / $0.40 |
google/gemini-2.5-flash-lite-batchbatch | vercel-gateway | $0.05 / $0.20 |
gemini-2.5-flash-litebatch | gemini API 경로를 사용할 수 없음 | $0.15 (Input price (text, image, video); (audio)); $0.05 (Input price (text, image, video); (text / image / video)) / $0.20 |
gemini-2.5-flash-liteflex | gemini API 경로를 사용할 수 없음 | $0.15 (Input price (text, image, video); (audio)); $0.05 (Input price (text, image, video); (text / image / video)) / $0.20 |
gemini-2.5-flash-litepriority | gemini API 경로를 사용할 수 없음 | $0.54 (Input price (text, image, video); (audio)); $0.18 (Input price (text, image, video); (text / image / video)) / $0.72 |
gemini-2.5-flash-litestandard | gemini API 경로를 사용할 수 없음 | $0.30 (Input price (text, image, video); (audio)); $0.10 (Input price (text, image, video); (text / image / video)) / $0.40 |
gemini-2.5-flash-lite · batch · gemini · Context caching price: $1.00 / million token hours · / 1,000,000 tokens per hour (storage price)gemini-2.5-flash-lite · flex · gemini · Context caching price: $1.00 / million token hours · / 1,000,000 tokens per hour (storage price)gemini-2.5-flash-lite · priority · gemini · Context caching price: $1.80 / million token hours · / 1,000,000 tokens per hour (storage price)gemini-2.5-flash-lite · standard · gemini · Context caching price: $1.00 / million token hours · / 1,000,000 tokens per hour (storage price)Fix default Gemini model: gemini-2.5-flash-lite is retired for new accounts (404), use gemini-3.5-flash
on Gemini's free tier, so after its daily `gemini-2.5-flash-lite` quota ran out, the rest of the run used `gemini-3.1-flash-lite` (same code path). Production's configured model
- Default model gemini-2.5-flash-lite (--model overrides; flash is opt-in).
- **Default model `gemini-2.5-flash-lite`** (`--model` / `TRANSLATE_MODEL` override; flash is opt-in). Workflow default updated too.
its model; with no profile selected it reads `google/gemini-2.5-flash-lite · default`.
fix: update gemini-2.5-flash-lite to gemini-3.5-flash-lite in config/models.yaml
- Model list was stale. gemini-2.5-flash-lite led the list, but Google now
allowlist (Workers AI `@cf/meta/llama-3.1-8b-instruct-fp8` primary, `@cf/qwen/qwen3-30b-a3b-fp8` alternate; conditional `gemini-2.5-flash-lite` fallback marked `unverified_free_tier` — stays OFF), validated bounds (≤16 KiB input / ≤768 output tokens /
Gemini 2.5 Flash-Lite: Google의 AI 모델, 출시일 2025-07-22. 컨텍스트 윈도는 1,048,576 토큰입니다. 최대 출력은 65,536 토큰입니다. 토큰 100만 개당 요금은 입력 $0.10, 출력 $0.40입니다.
컨텍스트 윈도는 1,048,576 토큰입니다. 최대 출력은 65,536 토큰입니다.
토큰 100만 개당 요금은 입력 $0.10, 출력 $0.40입니다. 캐시 입력은 100만 개당 $0.01입니다.
2025-07-22.