Tencent Hy4 Flash
Summary
Tencent Hy4 Flash is an open-weight language model from Tencent (Hunyuan), released on 16 Nov 2026. It accepts text and images and generates text, with a 1M-token context window and up to 64K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.04 per million input tokens and $0.14 per million output tokens (OpenRouter), across 3 providers we track.
Tencent's lower-cost Hy4 tier for everyday agent and productivity tasks, with image understanding and a 1M-token context window.
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Tencent (Hunyuan) |
|---|---|
| Release date | 16 Nov 2026 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | tencent-hy-license [Hugging Face] |
| Parameters | 220B [Hugging Face] |
| Context window | 1M (1,024,000 tokens) [models.dev] |
| Max output | 64K tokens [models.dev] |
| Input | Text and images [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Hugging Face | tencent/Hy4-flash [Hugging Face] |
Tencent Hy4 Flash API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| OpenRouter tencent/hy4-flash | $0.04 | $0.14 | $0.0015 | 1M | OpenRouter, 18 Nov 2026 |
| Vercel AI Gateway tencent/hy4-flash | $0.04 | $0.14 | $0.0015 | 1M | models.dev, 18 Nov 2026 |
| Novita AI tencent/hy4-flash | $0.07 | $0.28 | $0.003 | 1M | models.dev, 18 Nov 2026 |
Running Tencent Hy4 Flash locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 134 GB | 134 GB | 135 GB |
| Q8_0 (8-bit) | 234 GB | 235 GB | 235 GB |
| FP16 / BF16 (16-bit) | 441 GB | 441 GB | 442 GB |
At 4-bit with an 8K conversation, Tencent Hy4 Flash needs about 134 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of Tencent Hy4 Flash?
Tencent Hy4 Flash has a context window of 1,024,000 tokens (1M) and can generate up to 64,000 tokens in a single response, according to models.dev.
How much does the Tencent Hy4 Flash API cost?
Tencent Hy4 Flash is listed from $0.04 per million input tokens and $0.14 per million output tokens (OpenRouter). Prices last verified 18 Nov 2026.
Is Tencent Hy4 Flash open source?
Tencent Hy4 Flash is an open-weight model: its weights are published on Hugging Face as tencent/Hy4-flash under the tencent-hy-license license. Check the license terms before commercial use.
When was Tencent Hy4 Flash released?
Tencent (Hunyuan) released Tencent Hy4 Flash on 16 Nov 2026, according to models.dev.
Which providers offer Tencent Hy4 Flash?
We track Tencent Hy4 Flash on 3 providers: OpenRouter, Vercel AI Gateway, Novita AI.
How much VRAM does Tencent Hy4 Flash need?
About 134 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 235 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | DeepSeek | 10 Sept 2026 | 1M | $0.05 | $0.20 | Open |
| ricursiveAI/Ripple-65B-Preview | Ricursiveai | 4 Nov 2026 | 1M | $0.10 | $0.30 | Open |
| MiniMax-M3.1-Flash | MiniMax | 19 Oct 2026 | 1M | $0.07 | $0.30 | Open |
| DeepSeek V4 Flash 0731 | DeepSeek | 31 Jul 2026 | 1M | $0.06 | $0.18 | Open |
| Muse Spark 1.3 Contributor | Meta | 2 Sept 2026 | 1M | $0.10 | $0.20 | – |
| Muse Spark 1.2 Contributor | Meta | 21 Aug 2026 | 1M | $0.10 | $0.20 | – |
Change history
- Open weights changed from no to yes
- Parameter count changed from none to 220B
- License changed from none to tencent-hy-license
- Now listed on Novita AI
- Now listed on OpenRouter
- Added to the directory
- Now listed on Vercel AI Gateway