- Size
- 220B parameterscounted from the safetensors header on the Hub
- Licence
- custom, read it, Tencent Hy License, Version 1.0Our reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
- On the Hub since
- 17 Nov 2026the date the repo was created
- Context length
- 1M tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)HYV4FlashForConditionalGeneration
- On Hugging Face
- tencent/Hy4-flash2.1k downloads in the last 30 days, 292 likes
Scores
No independent scores for this model in Epoch AI's data. A score is only shown here when Epoch's name for a model matches this repo with certainty; an uncertain match is left out rather than guessed.
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.11 GB, worked out from the model's config (tencent/Hy4-flash), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
GGUF for llama.cpp, Ollama, LM Studio and others
AtomicChat/Hy4-flash-GGUF (converted by AtomicChat), 4.2k downloads. Sizes are the real files on the Hub, all parts added up. For image input you also need the projector file mmproj-Hy4-flash-F16.gguf (0.91 GB), not counted below.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| IQ2_XXS | 78.9 GBfile size, 2 parts | 79.5 GB78.9 + 0.11 + 0.50 | Commandsllama.cpp llama-server -hf AtomicChat/Hy4-flash-GGUF:IQ2_XXSDownload hf download AtomicChat/Hy4-flash-GGUF --include "Hy4-flash-AD-IQ2_S-IQ2_XXS-*-of-00002.gguf" --local-dir Hy4-flash-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 4bitTensorFold/Hy4-flash-MLX-4bit-MTP | 120.9 GBfile size | 121.6 GB120.9 + 0.11 + 0.50 | Download page |
| MLX bf16apetersson/Hy4-flash-oQh8-BF16-Engram | 151.5 GBfile size | 152.1 GB151.5 + 0.11 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 410.3 GBfile size | 410.9 GB410.3 + 0.11 + 0.50 |
hf download tencent/Hy4-flash --local-dir Hy4-flashOther conversions (9; sizes not read)
- tencent/Hy4-flash-FP8 FP8, by the maker, 766 downloads
- AngelSlim/Hy4-flash-GGUF GGUF, 56k downloads; sizes not used: file sizes do not match this model's 220.3B parameters (median 1.61 times the nominal size); the repo may be tagged wrongly
- Avifenesh/Hy4-flash-NVFP4 NVFP4, 381 downloads
- LibertAIDAI/Hy4-flash-EXL3-4bpw EXL3, 83 downloads
- LibertAIDAI/Hy4-flash-EXL3-3.5bpw EXL3, 53 downloads
- leoncca/Hy4-flash-AWQ-g32 AWQ, 36 downloads
- LibertAIDAI/Hy4-flash-EXL3-3bpw EXL3, 35 downloads
- LibertAIDAI/Hy4-flash-EXL3-2.5bpw EXL3, 9 downloads
- infosave/Hy4-flash-cmf Other, 5 downloads
The Hub tags 22 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
1 provider serves this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.
| Provider | In | Out | Context | Precision | Tokens/s | First token | Uptime 24h | Tools | JSON |
|---|---|---|---|---|---|---|---|---|---|
| Tencentthe maker | $0.035 | $0.14 | 1M | fp8 | 65 | 3.0 s | 100.0% | yes | yes |
Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.
curl https://router.huggingface.co/v1/chat/completions \
-H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \
-d '{"model": "tencent/Hy4-flash:tencent", "messages": [{"role": "user", "content": "Hello"}]}'Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.