The whole dayWed 18 Nov 2026
AI Compass
WeeklyArchiveAbout

Tencent releases open-weight 220B Hy4-flash models with 1M context

Instruct, FP8 and pretrained base checkpoints with DFly2 speculative decoding weights under the Tencent Hy License across Hugging Face, ModelScope and GitCode on 18 November 2026.

Glowing cobalt-and-ice-blue Tencent Hunyuan swirl sphere above a silicon wafer routing cyan optical beams through a crystal KV-cache cube into three open-weight GPU modules
Image: AI Compass illustration (Tencent Hunyuan marks)

Two days after launching its hosted Hy4 flash API, Tencent Hunyuan released open weights on 18 November 2026 for tencent/Hy4-flash, tencent/Hy4-flash-FP8 and the pretrained tencent/Hy4-flash-base checkpoint across Hugging Face, ModelScope and GitCode under the Tencent Hy License. Listed on the Hugging Face Hub at 220 billion total parameters including 64 billion Engram lookup parameters (224 billion in Tencent’s earlier research note) with 12 billion active parameters per token, the natively multimodal Mixture-of-Experts model accepts text and image inputs across a 1-million-token context window.

Open Benchmarks scatter plot of score versus cost per task in RMB showing Hy4 flash on the left Pareto frontier ahead of Qwen3.8-Flash-Next, GLM-5.3-Flash, GPT-6 Luna and Hy4 preview
Across seven public agent benchmarks, Hy4-flash leads the low-cost frontier ahead of Qwen3.8-Flash-Next and GPT-6 Luna.Image: Tencent

LatentMoE KV-cache compression, vLLM and SGLang serving, and DFly2 decoding

Hy4-flash pairs hybrid attention, LatentMoE and Engram tables to compress KV-cache memory to one-sixth of flagship Hy4 while halving average output token length through self-distillation and reinforcement learning. Alongside the full-precision and FP8 checkpoints, Tencent released DFly2 speculative decoding weights that reuse target top-k sparse KV states and hidden-layer corrections across draft blocks.

DFly2 architecture diagram showing target-model DSA layer KV Share and Index Share feeding three draft Transformer blocks and a Hidden-Correction head
Tencent’s companion DFly2 speculative decoder reuses target top-k sparse KV caches and last-layer hidden states via gated fusion.Image: Tencent

Developers can run the weights locally or on private clusters through Hugging Face Transformers (image-text-to-text), vLLM, SGLang, Docker Model Runner (hf.co/tencent/Hy4-flash) and community GGUF quantisations for llama.cpp, Ollama and LM Studio.

Technical details

Released Hugging Face repos
tencent/Hy4-flash, tencent/Hy4-flash-FP8 and tencent/Hy4-flash-base (plus DFly2 speculative decoding)
Parameters and context
220B listed on Hub (including 64B Engram), 12B active per token, 1M text-and-image context window
Licence and distribution
Tencent Hy License across Hugging Face, ModelScope and GitCode
Supported serving engines
Transformers, vLLM, SGLang, Docker Model Runner and GGUF (llama.cpp, Ollama and LM Studio)

Release details & context

Seven-benchmark public evaluation suite and DFly2 sparse KV reuse

On the Hugging Face model card, Tencent’s Open Benchmarks frontier aggregates seven external agentic and reasoning evaluations: DeepSWE, ProgramBench, Terminal-Bench 4.0, APEX-Agents, AutomationBench, Humanity’s Last Exam (HLE) and OSWorld 2.1, plotted alongside internal eight-track WorkIndex and 259-task blind WorkBuddy evaluations.

To sustain high single-stream decoding throughput on self-hosted clusters, the DFly2 architecture shares top-k sparse KV tensors and DSA indexer states directly from frozen target layers into three lightweight draft Transformer blocks, applying gated fusion with the last accepted hidden state and a top Hidden-Correction layer.

Common questions

Which checkpoints are included in Tencent’s Hy4-flash open-weight release?

Tencent published the post-trained tencent/Hy4-flash instruct model, the quantised tencent/Hy4-flash-FP8 checkpoint and the pretrained tencent/Hy4-flash-base model alongside DFly2 speculative decoding support.

Sources

  1. Primarytencent/Hy4-flash on Hugging Facehuggingface.co/tencent
  2. Officialtencent/Hy4-flash-FP8 on Hugging Facehuggingface.co/tencent
  3. Officialtencent/Hy4-flash-base on Hugging Facehuggingface.co/tencent

Tencent release timeline (3 launches tracked)

  1. Earlier4 Nov 2026
    Tencent launches multimodal Hy4 model with native computer use and 3D engine training

    Tencent released Hy4 on Tencent Cloud TokenHub at roughly $0.90 per 1M input tokens, adding multimodal perception, computer use and 3D engine workflows.

  2. Earlier16 Nov 2026
    Tencent launches 224B Hy4 flash multimodal MoE at $0.070 input

    Tencent launches its 224B-total, 12B-active Hy4 flash multimodal model at $0.070 input and $0.281 output, at one-twentieth of Hy4’s task cost.

Explore the full Tencent dossier or browse all open source releases.

Also in the Wednesday 18 November edition

Read all 18 launches on one page
  1. 01
  2. 02
  3. 03
  4. 04
  5. 05
  6. 06
  7. 08
  8. 09
  9. 10
  10. 11
  11. 12
  12. 13
  13. 14
  14. 15
  15. 16
  16. 17
  17. 18

Spotted an error? Email aiforsuccess.in+compass@gmail.com or tag @ai_for_success on X