Xylolabs

Hy4 flash GGUF

Base model Formats License Code

GGUF quants of tencent/Hy4-flash (220B parameters, 12B active) for llama.cpp, on CUDA, Metal or CPU. The Engram tables sit in a separate file that llama.cpp reads from disk as needed, so only the trunk has to fit in memory.

  • Files: quants from Q8_0 down to IQ2_M, the vision tower and the importance matrix.
  • Q4_K_M: KL 0.244 from BF16, top-1 85.1 %.

Files

Each quant has two files:

  • Hy4-flash-<QUANT>-00001-of-00002.gguf is the trunk.
  • Hy4-flash-<QUANT>-00002-of-00002.gguf holds the Engram tables (63.4 GB, FP8 E4M3). It is the same file for every quant (sha256 c3589e8789026f7de689a5bd2258daf018deddcbd903a443713051a6d84351a9), and llama.cpp picks it up from the same folder.

Also in this repository: mmproj-Hy4-flash-BF16.gguf, the vision tower (1.04 GB), and Hy4-flash.imatrix.gguf, the importance matrix.

Quant Shard 1 (trunk) PPL Mean KL vs BF16 Same top-1 vs BF16
Q8_0 165.50 GB 29.60 0.1741 87.9%
Q6_K 129.89 GB 28.21 0.1930 87.2%
Q5_K_M 113.37 GB 29.19 0.1868 86.9%
Q4_K_M 97.81 GB 30.96 0.2441 85.1%
Q4_K_S 92.19 GB 31.83 0.2660 84.8%
IQ4_XS 86.94 GB 30.27 0.3027 83.5%
Q3_K_M 78.58 GB 27.41 0.4046 80.7%
IQ3_M 72.49 GB 37.28 0.4839 78.3%
IQ3_XXS 64.89 GB 31.47 0.5165 77.3%
Q2_K 61.79 GB 29.48 0.7075 73.1%
IQ2_M 56.26 GB 33.49 0.7107 72.9%

How we measured:

  • KL and top-1: against the BF16 model (the BF16 trunk with the same Engram file, PPL 30.39), on 24 chunks of 512 tokens of wikitext-2.

Mean KL and same top-1 against BF16 by trunk size

Sizes are the trunk only.

Q8_0, Q6_K, Q5_K_M, Q4_K_M and Q4_K_S are quantized from the BF16 original, tencent/Hy4-flash. The other files were quantized from tencent/Hy4-flash-FP8, Tencent's FP8 release.

Compared with Tencent's Q4_K_M

Q4_K_M from Trunk size PPL Mean KL vs BF16 Same top-1 vs BF16
Xylolabs, from BF16 with an F16 LM head (this repository) 97.81 GB 30.96 0.2441 85.05%
AngelSlim (Tencent) 104.99 GB 31.92 0.2814 84.44%

Both were measured against the same BF16 reference, build and text; perplexity on raw text doesn't rank quants of this model reliably, so compare KL and top-1.

Tensor recipe

Tensor class Format
Routed experts ffn_gate_exps, ffn_up_exps, ffn_down_exps (148.6B parameters) the quant type of the file
Attention (MLA projections, gate, output), shared experts, latent projections, layer-0 dense MLP, token embedding, indexer projections, Engram key/value projections Q8_0
Router ffn_gate_inp, hyper-connection functions, indexer proj and compressor, small Engram weights, norms, biases F32
LM head output.weight F16 (Q8_0 in the Q8_0 file)
Engram tables per_layer_token_embd (63.4B parameters) FP8 E4M3

How to run

Hy4-flash support in llama.cpp is pull request #32517 (commit 0a054d0b0) until it is merged. The same build runs on CUDA and on Metal. Tencent's patched build also loads these files.

git clone https://github.com/ggml-org/llama.cpp && cd llama.cpp
git fetch origin pull/32517/head:hy4-flash && git checkout hy4-flash
cmake -B build -DCMAKE_BUILD_TYPE=Release        # add -DGGML_CUDA=ON for NVIDIA GPUs
cmake --build build -j --target llama-cli llama-server

./build/bin/llama-cli -m Hy4-flash-Q4_K_M-00001-of-00002.gguf \
  --override-tensor '^per_layer_token_embd\.weight$=CPU' --load-mode mmap --lazy-mode on \
  -ngl 99 --jinja -c 8192

--override-tensor keeps the Engram tables in host memory, and --load-mode mmap --lazy-mode on reads them from disk only when a prompt needs them. llama-server takes the same flags.

How these were made

  1. Convert with the PR's convert_hf_to_gguf.py plus hy4-flash-release-convert.patch (in this repository): the trunk in bf16, the Engram file, and the vision tower with --mmproj.
  2. Build the importance matrix with llama-imatrix on the Q8_0 trunk, 125 chunks of 512 tokens.
  3. Quantize: llama-quantize --imatrix Hy4-flash.imatrix.gguf --output-tensor-type bf16 --tensor-type-file tensor_types_base.txt <bf16 trunk> <out> <QUANT>.
  4. Split each file into its two parts with llama-gguf-split.

License and notices

  • License: Tencent Hy License, Version 1.0, copied in LICENSE. Its terms apply to these files, including the extra terms for large Model-as-a-Service operators.
  • What we changed: we quantized the weights and converted them to GGUF. The conversion patch is hy4-flash-release-convert.patch.
  • Notices: NOTICE names the source model and its copyright holder.
  • Xylolabs is not affiliated with or endorsed by Tencent.

Tencent's original model description

The text below is reproduced from the README of tencent/Hy4-flash (the model introduction, highlights, benchmarks, quickstart and deployment sections), copyright Tencent, under the Tencent Hy License (see above). It describes Tencent's original model, not these files, and its usage instructions (Transformers, vLLM, SGLang) are for Tencent's checkpoints; for the GGUF files use "How to run" above. Changes by Xylolabs: heading levels are one level deeper, and image and link paths point to Tencent's repository instead of copies; the wording is unchanged. Tencent's own README is the authoritative version.

Model Introduction

Hy4 flash is our most cost-effective Hy model to date, featuring 220B total parameters (including 64B Engram parameters) with just 12B active parameters per token. It natively accepts text and image inputs, generates text, and supports a context window of up to 1M tokens.

Its 42-layer language backbone consists of one dense feed-forward layer and 41 MoE layers, with identity Hyper-Connections (iHC) supporting inter-layer information flow. A vision encoder (adapted from SigLIP2-So400m-Patch16-Res512) is connected to the backbone through an MLP projector. In addition to the backbone, a built-in 5-layer drafter, DFly2 (0.85B parameters), reuses the target model's KV cache and sparse indices to support long-context speculative decoding at a low, constant cost.

Compared with Hy4, Hy4 Flash introduces three notable architectural changes:

  1. Hybrid attention. Sparse attention (IndexPool) and sliding-window attention (SWA) are interleaved at a 1:3 ratio.
  2. LatentMoE. Routed experts operate in a lower-dimensional latent space. Each token activates 16 of 512 routed experts, alongside one shared expert.
  3. N-gram embedding (Engram). The backbone is augmented with learned bigram and trigram representations, retrieved via multi-head hash lookups.
Property Value
Overview
Architecture Mixture-of-Experts (MoE)
Total Parameters (excluding Engram, ViT, and DFly2) 155.5B
Engram Parameters 63.5B
ViT Parameters 0.43B
DFly2 Parameters 0.85B
Activated Parameters per Token Text 11.3B + Vision 0.43B
Context Length 1M (1,048,576 tokens)
Language Backbone
Backbone Layers 42 — 1 dense layer and 41 MoE layers
Hidden Size 3,072
Vocabulary Size 120,832
Residual Connections iHC (identity Hyper-Connections) with 4 residual streams
Attention — Hybrid Layout
Attention Architecture Sparse attention + sliding-window attention (SWA)
Backbone Layer Pattern 2 × Sparse → 10 × (3 × SWA → Sparse)
Attention Formulation Multi-head Latent Attention (MLA)
Multimodal Position Encoding Interleaved MRoPE
Attention — Layers Sparse / SWA
Backbone Layers 12 / 30
Attention Heads 32 / 64
MLA q_lora_rank 1,024 / 1,024
MLA kv_lora_rank 512 / 512
QK Head Dimension (non-RoPE + RoPE) 192 + 64 / 192 + 64
Value Head Dimension 256 / 256
RoPE Base 10,000,000 / 10,000
Sparse Context Selection Up to 2,048 tokens via group-level selection
SWA Context Window 512-token causal window (511 preceding + current)
Feed-Forward Networks
Dense FFN Intermediate Size 13,824
Routed Experts per MoE Layer 512
Shared Experts per MoE Layer 1
Activated Routed Experts per Token 16
MoE Latent Dimension 1,536
Expert Intermediate Size 1,536
N-gram Embedding
N-gram Orders Bigrams and trigrams
Backbone Layer Indices (starting from 0) 1, 11, 21, 31
Hash Heads per N-gram Order 8
Hash Vocabulary Size per Head 10.3M
Embedding Dimension per N-gram Order 768 (8 heads × 96 dimensions)
Embedding Table Parameters 63.4B
Vision Encoder
Vision Encoder Layers 27
Vision Encoder Hidden Size 1,152
Vision Encoder Intermediate Size 4,304
Vision Encoder Attention Heads 16
Vision Patch Size 16 × 16
Vision Position Embedding Size 128 × 128
Visual Feature Merging 2 x 2

Pretrained model capability

Bits-per-byte (BPB) on internal code & agent validation, lower is better:

Bits-per-byte on internal code and agent validation versus active parameters; lower is better

Few-shot pretrain benchmark evaluation:

Category Benchmark #Shots Hy4 flash Base Hy4 Base Hy3 preview Base DeepSeek-V4-flash-Base DeepSeek-V4-Pro-Base
Knowledge MMLU† 5 88.7 91.6 87.5 89.7 91.1
MMLU-Pro† 5 71.1 73.9 65.8 68.6 73.4
MMLU-Redux† 5 87.6 90.6 87.0 88.8 90.2
SuperGPQA† 5 57.8 62.7 52.3 52.5 54.8
C-Eval† 5 92.0 95.0 90.1 91.9 93.4
CMMLU† 5 91.6 93.1 89.6 89.8 90.7
SimpleQA 5 37.5 57.2 26.4 31.3 52.1
Chinese-SimpleQA† 5 74.6 83.9 69.6 71.9 76.6
AGIEval (English)† 3 75.4 79.3 76.0 76.3 78.2
AGIEval (Chinese)† 3 90.6 91.2 87.8 88.5 89.8
Reasoning HellaSwag (3,000-item subset) 10 85.2 87.1 85.7 85.1 86.8
PIQA 4 84.4 85.3 84.6 83.8 84.9
ARC-Challenge 0 95.0 97.3 96.0 97.3 97.0
Social IQa 0 80.1 82.7 80.5 72.5 73.1
KOR-Bench 3 54.7 57.6 54.6 53.4 55.4
C³ 0 77.3 82.4 75.9 74.1 78.3
Math U-MATH 5 69.8 71.6 60.9 57.4 64.8
OCW Courses 5 64.0 66.7 62.6 63.3 68.2
Code MBPP-plus 3 73.7 74.0 77.4 76.8 79.1
Multilingual Belebele (5-language subset)† 5 93.8 95.0 94.2 94.4 95.2
INCLUDE† 5 82.8 88.1 84.3 84.0 84.3
Multimodal MMMU-Pro (Standard, 10 opts) 4 52.0 59.6 — — —
SimpleVQA 4 48.5 54.5 — — —
CVBench 4 83.4 86.7 — — —
RealWorldQA 4 71.1 72.3 — — —
TextVQA 4 73.4 78.2 — — —
CountBenchQA 4 86.6 90.6 — — —

† Macro-averaged accuracy for the marked text benchmarks. Other text rows use the accuracy metric reported by our evaluation suite. Bold marks the highest available numerical score in each row; it does not indicate statistical significance.

Multimodal results are reported only for Hy4 flash Base and Hy4 Base.

ARC-Challenge and HellaSwag use 299-item and 3,000-item subsets, respectively. INCLUDE uses a 5,000-example Lite subset covering 44 languages. Belebele covers German, Korean, Vietnamese, Spanish, and Russian (4,475 examples in total).

Disclaimer: Because our base model was exposed to a large amount of synthetic instruction data during mid-training, it occasionally fails to follow the expected answer format in few-shot evaluations, which may distort the scores.

Practicality and Cost Efficiency

Optimization Approach

Across training, evaluation, and inference, we aggressively optimized for cost efficiency while preserving intelligence:

  • Pre-training: Hy4 flash integrates hybrid attention, LatentMoE, and Engram into its architecture—reducing KV cache size to just 1/24 of Hy3 and 1/6 of Hy4, with extensive scaling ladder experiments ensuring performance meets expectations.
  • Post-training: Self-distillation from Hy4 establishes strong general intelligence, while reinforcement learning cuts average end-to-end output tokens by 50%.
  • Evaluation: We built WorkIndex, an internal benchmark grounded in typical WorkBuddy tasks, and paired it with human evaluation—focusing on deliverable quality, single-turn output length, and task turnaround time rather than public benchmark scores detached from everyday work.
  • Inference: Combining megakernels, 4-bit quantization, and DFly2 pushes decoding speed past 150 tokens/s, enabling low-cost serving at scale.

Task Performance and Cost

Hy4 flash delivers overall task performance slightly above Hy4 preview at a much lower cost per task. On WorkIndex, it scores 42.1, compared with 39.7 for Hy4 preview, while its average cost per task is just 1/13 of Hy4 preview, 1/20 of Hy4, and 1/8 of DeepSeek-V4.1-Flash-1017 (Max tier, peak-hour pricing).

Performance and cost per task on WorkIndex for Hy4 flash and comparison models Performance and cost per task on Open Benchmarks for Hy4 flash and comparison models

Figure 1: Performance and cost across internal and external tasks. WorkIndex (internal) comprises 8 evaluation sets covering slide creation, slide editing, human-AI slide co-writing, Markdown human-AI co-writing, document editing, spreadsheet repair, audio/video processing, and visual understanding.
Open Benchmarks includes 7 public agent benchmarks: DeepSWE, ProgramBench, Terminal-Bench 4.0, APEX-Agents, AutomationBench, HLE, and OSWorld 2.1.

Average cost per task on WorkIndex for Hy4 flash, Hy4 preview, and Hy4

Figure 2: Average cost per task on WorkIndex. Task scores are 42.1 for Hy4 flash, 39.7 for Hy4 preview, and 51.4 for Hy4.

We recommend Hy4 flash for most everyday productivity tasks, including deep search, document processing, slide creation, simple frontend demos, and computer use. For complex tasks such as long-horizon software engineering, we recommend Hy4 for its higher intelligence ceiling.

User Blind Evaluation

In the blind evaluation, Hy4 flash achieved user satisfaction close to Claude Haiku 5.5, while recording the lowest cost per task and shortest median end-to-end completion time among the evaluated models. Its user satisfaction score was 2.79/4 (Claude Haiku 5.5: 2.81/4), with a median end-to-end completion time of 6.7 minutes and a cost of RMB 0.26 per task.

Blind evaluation comparing user satisfaction, median end-to-end completion time, and cost per task

Figure 3: Blind evaluation with 85 internal WorkBuddy users across 259 tasks spanning 27 categories.

Appendix: Benchmark results

Full benchmark scores, task costs, and evaluation notes for Hy4 flash and comparison models

Quickstart

Use Transformers to load the model for local inference, or deploy a server and access it through the OpenAI-compatible API.

Reasoning mode (instruct models): reasoning_effort supports "high" (default, thinking enabled) and "no_think" (direct responses). Both examples below use "no_think".

Recommended sampling parameters: temperature=0.9, top_p=1.0.

Transformers

Support for Hy4 flash will be merged into the transformers main branch soon. Until then, install from the commit below.

pip install "transformers[torch,vision] @ git+https://github.com/huggingface/transformers.git@refs/pull/50903/head"

The following example generates a brief image description:

from transformers import AutoModelForImageTextToText, AutoProcessor

model_id = "tencent/Hy4-flash"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(model_id, device_map="auto")

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
            },
            {"type": "text", "text": "Describe this image in one sentence."},
        ],
    }
]

inputs = processor.apply_chat_template(
    messages,
    add_generation_prompt=True,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    reasoning_effort="no_think",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0, inputs["input_ids"].shape[1] :], skip_special_tokens=True))

OpenAI-compatible API

For API access, start a server using the vLLM deployment instructions, then run the following client example:

from openai import OpenAI

client = OpenAI(base_url="http://127.0.0.1:8000/v1", api_key="EMPTY")

response = client.chat.completions.create(
    model="hy4-flash",
    messages=[
        {"role": "user", "content": "Hello! Can you briefly introduce yourself?"},
    ],
    temperature=0.9,
    top_p=1.0,
    extra_body={"chat_template_kwargs": {"reasoning_effort": "no_think"}},
)
print(response.choices[0].message.content)

Deployment

DFly2 for Speculative Decoding

DFly2 is Hy4 flash's speculative-decoding drafter, trained with AngelSpec and built directly on the target model's Top-k Sparse KV: each draft layer shares the KV cache (KV Share) and the compressed indexer's selection of up to 2,048 positions (Index Share) with its corresponding sparse attention layer. This yields three advantages:

  • Constant cost: each draft token attends to at most ~2K keys, regardless of context length.
  • Zero extra KV memory: no separate KV cache is maintained for the drafter.
  • PD-disaggregation ready: not involved in prefill; the prefill-to-decode transfer adds only one last-layer hidden state.

DFly2 architecture: each draft layer reuses the KV cache and the indexer-selected positions of a sparse attention layer of Hy4 flash

Quantization

We provide AngelSlim, a more accessible, comprehensive, and efficient toolkit for large model compression. AngelSlim supports a comprehensive suite of compression tools for large-scale multimodal models, including common quantization algorithms, low-bit quantization, and speculative sampling.

For this model, we have provided a GGUF compressed version for end-to-end testing purposes.

Inference Frameworks

For production serving, we recommend using vLLM or SGLang. Please refer to the recipes:

vLLM

Use official prebuilt image vllm/vllm-openai:hy4-flash:

docker run --gpus all \
  -p 8000:8000 \
  --ipc=host \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  vllm/vllm-openai:hy4-flash tencent/Hy4-flash-FP8 \
    --tensor-parallel-size 8 \
    --speculative-config '{"num_speculative_tokens":5,"method":"dfly"}' \
    --attention-backend FLASHMLA_SPARSE \
    --tool-call-parser hy_v4 \
    --reasoning-parser hy_v4 \
    --enable-auto-tool-choice \
    --port 8000 \
    --served-model-name hy4-flash
SGLang

Use the official prebuilt image lmsysorg/sglang:hy4-flash (multi-arch, x86 and Arm):

docker pull lmsysorg/sglang:hy4-flash

docker run --gpus all \
  -p 8000:8000 \
  --ipc=host \
  lmsysorg/sglang:hy4-flash \
  sglang serve \
    --trust-remote-code \
    --model-path tencent/Hy4-flash-FP8 \
    --tp 8 \
    --cuda-graph-backend-prefill disabled \
    --speculative-algorithm DFLY \
    --speculative-dfly-block-size 6 \
    --host 127.0.0.1 \
    --port 8000
Downloads last month
2,747
GGUF
Model size
219B params
Architecture
hy_v4_flash
Hardware compatibility
Log In to add your hardware

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Xylolabs/Hy4-flash-GGUF

Quantized
(22)
this model

Collection including Xylolabs/Hy4-flash-GGUF