# Qwen 3.7 Plus

> Alibaba's newest Plus flagship. 1M context, vision input, top agentic + reasoning.

Human version: https://kymaapi.com/models/qwen-3.7-plus
Live JSON: `GET https://kymaapi.com/v1/models` (no auth required)

## Facts

- **Model ID**: `qwen-3.7-plus` — pass this as `model` in the request body
- **Creator**: Alibaba
- **Released**: 2026-06-02
- **Context window**: 1M tokens
- **Max output**: 33K tokens per response — a hard ceiling, not a default
- **Price**: $0.4482 in / $1.793 out per 1M, cached input 10%
- **Capabilities**: tools, vision, reasoning, caching

## Call it

```bash
curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen-3.7-plus", "messages": [{"role": "user", "content": "Hello"}]}'
```

## Positioning

Qwen 3.7 Plus is a balanced-tier frontier model built for general-purpose tasks, agentic workflows, and long-context reasoning. Developers should reach for it when they need reliable tool use, image understanding, and structured outputs without premium-tier pricing.

## About

Alibaba's Qwen 3.7 Plus is a 1M-context language model that accepts text and image inputs. It supports tool calling, chain-of-thought reasoning, and structured JSON outputs, making it suitable for chat interfaces, code generation, and autonomous agent loops.

On Kyma, the model runs with automatic request failover and reports exact token costs in the usage.cost field. Prompt caching is enabled by default, billing repeated prompt prefixes at 10% of the standard input rate. The X-Kyma-Model header confirms which model handled each request.

Maximum output is capped at 32,768 tokens, and inference operates at a medium speed tier with a median throughput of 41 tokens per second. It does not accept audio or video inputs, and complex structured output schemas may introduce minor latency.

## Use cases

- **Agentic Workflow Orchestration** — Handles multi-step tool calls and state tracking for autonomous agent loops.
- **Long Context Document Analysis** — Processes large codebases or technical manuals within a 1M token window.
- **Vision Enabled Chat Interfaces** — Extracts and reasons over embedded images alongside conversational text.
- **Structured Data Extraction** — Outputs validated JSON for downstream API integration and database insertion.

## Not ideal for

Avoid this model if your application requires sub-20 token per second latency, native audio processing, or output lengths exceeding 32,768 tokens.

## Pick something else when

- You need faster inference for real-time chat → `gemini-3.5-flash-lite`
- You need maximum output length beyond 32k tokens → `qwen-3.7-max`, `deepseek-v4-pro`
- You want the lowest possible cost tier → `qwen3.7-flash`, `deepseek-v4-flash`

## See also

- All models: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md
- Other models by Alibaba: https://kymaapi.com/models?q=Alibaba