menu_bookDocumentation

Model Intelligence & Specs

A1Asset AI Studio dynamically routes your coding requests through high-performance language models optimized for speed and analytical reasoning.

DeepSeek V4 Flash 0731

Fast Response

Provider: OpenRouter

Context: 128,000 tokens

Max Output: 8192 tokens

A fast OpenRouter-hosted coding model tuned for responsive edits, explanations, and everyday implementation tasks.

Best for: Quick code changes, implementation Q&A, and rapid iteration.

GPT-4o mini

Fast Response

Provider: OpenAI

Context: 128,000 tokens

Max Output: 8192 tokens

A lightweight OpenAI model that balances low latency with reliable instruction following for common development tasks.

Best for: Explaining snippets, drafting small functions, and general coding assistance.

Qwen3.6 35B A3B

Fast Response

Provider: OpenRouter

Context: 128,000 tokens

Max Output: 8192 tokens

A fast Qwen model with strong coding coverage for short to medium implementation requests.

Best for: Code generation, syntax fixes, and focused refactoring.

Qwen3 Coder Plus

Complex Task

Provider: OpenRouter

Context: 128,000 tokens

Max Output: 8192 tokens

A complex-task Qwen coding model built for larger changes, planning, and multi-file reasoning.

Best for: Feature implementation, refactors, and agentic coding workflows.

Kimi K2.5

Complex Task

Provider: OpenRouter

Context: 128,000 tokens

Max Output: 8192 tokens

A complex-task model with strong long-context reasoning through OpenRouter.

Best for: Repository analysis, architecture changes, and deep debugging.

GLM-5.3-Flash

Fast Response

Provider: OpenRouter

Context: 128,000 tokens

Max Output: 8192 tokens

A natively multimodal Mixture-of-Experts (MoE) reasoning model from Zhipu AI with a 1M token context window and sparse/linear attention.

Best for: Analytical reasoning, high-efficiency multimodal coding, and long-context software engineering.

Gemini 3.5 Flash

Complex Task

Provider: Gemini

Context: 128,000 tokens

Max Output: 8192 tokens

A complex-task Gemini model for broad context processing and multimodal reasoning.

Best for: Complex debugging, multi-file code refactors, and architecture design.

Gemini 3.5 Flash-Lite

Fast Response

Provider: Gemini

Context: 128,000 tokens

Max Output: 8192 tokens

A speed-optimized Gemini model for low-latency everyday coding support.

Best for: Real-time chat, quick documentation lookups, and function refactoring.

Text-to-Image (gemini-3.1-flash-image-preview)

Complex Task

Provider: Google

Max Output: Media output

A media generation endpoint for text-to-image requests inside A1Asset AI Studio.

Best for: Generating visual assets, concepts, and image drafts.

Text-to-Video (veo-3.1-lite-generate-001)

Complex Task

Provider: Google

Max Output: Media output

A media generation endpoint for text-to-video requests inside A1Asset AI Studio.

Best for: Generating short video assets from text prompts.

Image-to-Video (veo-3.1-lite-generate-001)

Complex Task

Provider: Google

Max Output: Media output

A media generation endpoint for animating source images into video.

Best for: Image-to-video animation and visual motion drafts.

Subscription Quota Tiers

Your token usage is tracked automatically based on the complexity level of the model selected. Quotas refresh at the beginning of each billing cycle.

bolt

Fast Response Quota

30,000,000 Tokens / Month

Applies to active fast-response models: DeepSeek V4 Flash 0731, GPT-4o mini, Qwen3.6 35B A3B, GLM-5.3-Flash, Gemini 3.5 Flash-Lite.

hub

Complex Task Quota

1,000,000 Tokens / Month

Applies to active complex-task and media models: Qwen3 Coder Plus, Kimi K2.5, Gemini 3.5 Flash, Text-to-Image (gemini-3.1-flash-image-preview), Text-to-Video (veo-3.1-lite-generate-001), Image-to-Video (veo-3.1-lite-generate-001).