# Tencent Hy4 (slim)

> Tencent Hy4 is Tencent's flagship language model, released on 4 November 2026 with image understanding and computer use, served on Tencent Cloud TokenHub as hy4 with a 1M-token context, at ¥6 input and ¥18 output per million tokens.

- Full: https://www.anchorterminal.com/tools/hy4.md (~7,850 tokens) · this version ~1,330 tokens · JSON https://www.anchorterminal.com/tools/hy4.json · canonical https://www.anchorterminal.com/tools/hy4
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-13 · pricing https://www.anchorterminal.com/tools/hy4#pricing · reviews https://www.anchorterminal.com/tools/hy4#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**B · 64/100 · rank #595 of 1524 · #25 in Model APIs & inference · not agent-ready · confidence medium** · same model listed separately: Tencent Cloud TokenHub, Model router (https://www.anchorterminal.com/tools/tencent-tokenhub.min.md)


Assessment: Tencent's flagship model is on TokenHub as hy4, with a 1M-token context, image understanding, function calling and caching, at ¥6 input and ¥18 output per million tokens, half price until the end of 2026. The weights it is called open source for aren't published yet, and every benchmark is Tencent's own.

## Facts

- Kind: Model API · vendor: Tencent (Hy team) · category: Model APIs & inference · legal entity: Tencent Cloud Computing (Beijing) Co., Ltd. (腾讯云计算（北京）有限责任公司) · provenance 49/100
- Local only (HTTP)
- Auth: API key · pricing: Pay per use · x402: no · licence: Hosted model under the Tencent Cloud TokenHub terms. Tencent says weights and a system card will follow; no licence is stated yet
- Benchmark probes: not run yet, so Performance is pending
- Surface graded: hy4 on Tencent Cloud TokenHub, read from Tencent's release of 4 November and TokenHub's model list, price page and release notes on 11 November 2026. Platform readings (keys, SLA, terms, retention, errors, rate limits) are the tencent-tokenhub listing's of 1 November. No account was made and nothing was called
- Model ID: hy4. hy4-preview (28 August 2026) stays listed at the same price without the image feature
- Context: 1M tokens, 960K input and 64K output
- Capabilities: Thinking (preserved), structured output, function calling, caching and image understanding, per the model list. Computer use is in Tencent's release but not marked in the list
- Pricing: ¥6 input, ¥0.3 cached, ¥18 output per million tokens; 50% off until 31 December 2026
- Open weights: Promised with a system card 'in the coming days' on 4 November; not found on Hugging Face on 11 November. Hy4 preview's weights are Apache-2.0
- Tencent's figures: Vendor-reported: WorkBuddy benchmark average 83.18 and 96.15% pass rate; financial blind evaluation 3.04; GameDev up 19.1% and GameLogic up 9.8% on Hy4 preview; hallucination rate 7.2% (was 12.95%)
- Prices: Hy4 input $0.90 per 1M tokens; Hy4 cached input $0.045 per 1M tokens; Hy4 output $2.70 per 1M tokens

- Scores: Reliability 46, Performance pending, Schema & documentation 73, Agent ergonomics 86, Security & auth 71, Payments & pricing 45, Task success pending, Maintenance & community 73, Transparency & trust 52 · total over the 7 assessed categories
- Why: Reliability, Model reading for hy4 on Tencent Cloud TokenHub. · Schema & documentation, Model reading for hy4 on Tencent Cloud TokenHub. · Agent ergonomics, Model reading for hy4 on Tencent Cloud TokenHub. · Security & auth, Model reading for hy4 on Tencent Cloud TokenHub. · Payments & pricing, Model reading for hy4 on Tencent Cloud TokenHub. · Maintenance & community, Model reading for hy4 on Tencent Cloud TokenHub. · Transparency & trust, Model reading for hy4 on Tencent Cloud TokenHub.
- Sources: 4, open questions: 5, both in the full twin
- Capabilities: inference.llm
- JSON: https://www.anchorterminal.com/api/v1/tools/hy4.json
- Its vendor verifies the listing with `verify_listing` or `POST https://www.anchorterminal.com/api/v1/verify` (the full twin says how); no effect on the grade.
- Support: hunyuan_opensource@tencent.com; more in the full twin

## Before you call it

1. Use model id hy4 on TokenHub's OpenAI- or Anthropic-compatible endpoints; hy4-preview is a separate, older model
2. Keep input under 960K tokens and expect at most 64K output
3. Reuse prompt prefixes: cached input is ¥0.3 against ¥6
4. Check the price page for the end of the 50% discount on 31 December 2026 before budgeting
5. Read rate limits from the console or DescribeModelQuota; the public docs don't give them for chat models

## Connect

```bash
curl https://tokenhub.tencentmaas.com/v1/chat/completions \
  -H "Authorization: Bearer $TOKENHUB_API_KEY" -H "content-type: application/json" \
  -d '{"model":"hy4","messages":[{"role":"user","content":"hello"}]}'
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Vertex AI (Gemini Enterprise Agent Platform) | AA | 91.2 | inference.llm | https://www.anchorterminal.com/tools/vertex-ai.min.md |
| OpenAI API | AA | 85.8 | inference.llm | https://www.anchorterminal.com/tools/openai-api.min.md |
| Amazon Bedrock | A | 83.5 | inference.llm | https://www.anchorterminal.com/tools/amazon-bedrock.min.md |
| Azure OpenAI in Foundry Models | A | 81.2 | inference.llm | https://www.anchorterminal.com/tools/azure-openai.min.md |
| Claude API | A | 79.2 | inference.llm | https://www.anchorterminal.com/tools/anthropic-api.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)