Models (64)

Auto

DOS.AI Smart Router classifies each request by complexity. Simple and Medium requests use DOS.AI; Complex requests can use an eligible paid route configured in the live catalog.

DOS

DOS curated meta-model for DOSClaw and first-party DOS products.

Qwen3.8-27B

DOS.AI hosted Qwen3.8-27B for reasoning, coding, tool use, and multilingual text generation.

Granite 4.0 H Micro

Llama 3.2 1B Instruct

GPT-OSS 120B

OpenAI open-weight 120B, strong general reasoning at a fraction of hosted GPT pricing.

GPT-OSS 20B

OpenAI open-weight 20B, the cheapest capable chat model in the catalog.

Llama 3.2 3B Instruct

Qwen3 30B A3B FP8

GLM 4.7 Flash

Gemma 4 26B A4B

Tencent HY3

Xiaomi MiMo V2.5

GLM 5.3 Flash

Qwen3.8 Flash

Llama 3.1 8B

GPT-5.4 Nano

Cheapest GPT-5.4-class model for simple high-volume tasks

GPT-5.6 Luna

OpenAI most economical GPT-5.6 model, built for efficient high-volume workloads, with vision and a 1.05M context window.

Gemini 3.1 Flash-Lite

Fastest and most cost-efficient Gemini 3 model

Llama 3.3 70B

MiniMax M2.7

DeepSeek V4.1 Flash

Gemini 3.5 Flash-Lite

Fast, cost-effective Flash-Lite (GA). 1M context. Closest Gemini peer to dos-ai (Qwen3.6-35B-A3B). [PROMO] Free when used BY a DOSClaw agent on a paid plan (agent traffic only - direct API usage is billed normally). Limited-time.

Gemma SEA-LION v4 27B

Mistral Small 3.1 24B

Qwen 3.7 Plus

Qwen 3.7 Plus - multimodal (text+image), 1M context

DeepSeek R1 Distill Qwen 32B

Nemotron 3 120B A12B

NVIDIA Nemotron 3 Ultra

Qwen2.5 Coder 32B

QwQ 32B

Gemini 3.1 Flash Live

Real-time voice and dialogue model

Gemini 3.7 Flash

Google hybrid reasoning model combining fast multimodal inference with controllable thinking capabilities, a 1M context window, and up to 64K output tokens.

Gemini 3.8 Flash

Google's newest Gemini Flash model, with a 1M token context window.

GPT-5.4 Mini

Strong mini model for coding, computer use, and sub-agents

Kimi K2.6

Kimi K2.7 Code

Moonshot Kimi K2.7 Code, coding-focused with a 256K context window.

Claude Haiku 4.5

Fastest and most compact Claude model

Grok 4.3

xAI flagship reasoning model - 1M context, text+image (replaces grok-4.1-fast)

DeepSeek V4 Pro

Near-frontier quality at ~1/6 the cost of Opus 4.7 / GPT-5.5

GLM-5.2

Zhipu GLM-5.2 with a 1M context window, strong Chinese and English coding.

GLM 5.3

Gemini 3.5 Flash

Latest Gemini Flash - frontier performance, standard tier

Qwen 3.8 Max

Qwen 3.8 Max - Alibaba's 2.4T-parameter MoE flagship, 1M context, multimodal input (text+image+video)

Gemini 3.1 Pro

Google's most advanced reasoning model for complex tasks

GPT-5.6 Terra

OpenAI mid tier of the GPT-5.6 family, balancing intelligence and cost with vision, tool calling, and a 1.05M context window.

Grok 4.20

xAI flagship reasoning model with 2M context

Qwen 3.7 Max

Qwen 3.7 Max tier - 1M context (standard price; OpenRouter shows a 50%-off promo)

Kimi K3

Claude Sonnet 5

Anthropic balanced model with near-flagship coding and agentic quality at Sonnet pricing, a 1M context window, and up to 128K output tokens.

Claude Opus 5

Anthropic flagship model for complex agentic coding and enterprise work, with a 1M context window and up to 128K synchronous output tokens.

GPT-5.6 Sol

OpenAI frontier model for complex professional work, with maximum reasoning, vision, and a 1.05M context window.

Nano Banana 2 Lite

Google fast, low-cost image-output Gemini model. Output fixed at 1K resolution.

Nano Banana 2

Google mid-tier image-output Gemini model. Supports 0.5K/1K/2K/4K output resolutions (this catalog offering exposes up to 2K; see spec for 4K rationale).

Nano Banana Pro

Google premium image-output Gemini model. 1K and 2K are priced identically; this catalog offering exposes up to 2K.

Wan 2.2 T2I Flash

Alibaba DashScope fast text-to-image model (Wanx family). Submit+poll upstream, existing generateDashScopeImage pipeline, now reachable from the public /v1/images/generations route.

Wan 2.7 Image-to-Video

Video generation from image + text prompt via Alibaba Wan 2.7. Pricing: per 1000 seconds.

Wan 2.7 Text-to-Video

Video generation from text prompt via Alibaba Wan 2.7. Duration 2-15s, 1080P, native audio. Pricing: per 1000 seconds.

Qwen3 Embedding 4B

Self-hosted multilingual text embedding (2560 dims), strong on Vietnamese. Priced at our OpenRouter fallback cost.

Qwen3 Embedding 0.6B

Gemini Embedding 001

Google high-dimensional embedding model optimized for semantic search and Retrieval-Augmented Generation (RAG).

BGE Small EN v1.5

BGE Base EN v1.5

BGE Large EN v1.5