--- title: Qwen3.8-Flash-Next-NVFP4 canonical_url: "https://www.modelscope.cn/models/RadixArk/Qwen3.8-Flash-Next-NVFP4" md_url: "https://www.modelscope.cn/models/RadixArk/Qwen3.8-Flash-Next-NVFP4.md" repository: RadixArk/Qwen3.8-Flash-Next-NVFP4 last_updated: 2026-08-26 license: other pipeline_tag: image-text-to-text tasks: - image-text-to-text model_type: - qwen4_exp architectures: - Qwen4ExpForConditionalGeneration base_model: - Qwen/Qwen3.8-Flash-Next base_model_relation: quantized parameters: 119.6B tensor_type: - F32 - U8 - F8_E4M3 - BF16 - I64 library_name: - safetensors - pytorch frameworks: - pytorch downloads: 291080 stars: 9 tags: - ModelOpt - Qwen3.8 - quantized - FP4 - fp4 - sglang --- # Qwen3.8-Flash-Next-NVFP4 > Qwen3.8-Flash-Next-NVFP4 - RadixArk 在 ModelScope 开源的模型。Description: RadixArk Qwen3.8-Flash-Next-NVFP4 is the quantized version of Qwen/Qwen3.8-Flash-Next, a hybrid-architecture multimodal Mixture-of-Experts model. Quantization was performed with NVIDIA Model… RadixArk/Qwen3.8-Flash-Next-NVFP4 是 ModelScope 魔搭社区上的 119.6B 参数image-text-to-text模型，采用 other 许可，基于 Qwen/Qwen3.8-Flash-Next 构建。 - **Repository**: RadixArk/Qwen3.8-Flash-Next-NVFP4 - **License**: other - **Tasks**: image-text-to-text - **Parameters**: 119.6B - **Base model**: Qwen/Qwen3.8-Flash-Next - **Tags**: ModelOpt, Qwen3.8, quantized, FP4, fp4, sglang - **Downloads**: 291080 - **Stars**: 9 - **Last updated**: 2026-08-26 Source: https://www.modelscope.cn/models/RadixArk/Qwen3.8-Flash-Next-NVFP4 --- # Model Overview ## Description: RadixArk Qwen3.8-Flash-Next-NVFP4 is the quantized version of [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next), a hybrid-architecture multimodal Mixture-of-Experts model. Quantization was performed with [NVIDIA Model Optimizer](https://github.com/NVIDIA/TensorRT-Model-Optimizer) (snapshot `87c9f8cf`) using the NVFP4 W4A4 recipe, scoped to the routed experts only. This checkpoint is a private candidate release. ## References NVIDIA Model Optimizer: https://github.com/NVIDIA/TensorRT-Model-Optimizer ### License/Terms of Use: See the [source model](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) for license terms. ### Use Case: Developers evaluating NVFP4-quantized serving of Qwen3.8-Flash-Next for agentic systems, chat, coding, and multimodal reasoning workloads. ### Release Date: Hugging Face on 08/25/2026 via https://huggingface.co/RadixArk/Qwen3.8-Flash-Next-NVFP4 ## Model Architecture: **Architecture Type:** Transformer (hybrid GDN + QSA sparse attention, multi-hyperconnection streams, PLE n-gram injection)
**Network Architecture:** Multimodal MoE — 48 decoder layers, 512 routed experts per MoE layer (top-10 routing) + shared expert, 1 MTP layer
**Number of Model Parameters:** ~180B in total (360 GB BF16 source)
## Input: **Input Type(s):** Text, Image, Video
**Input Format(s):** String, RGB, Video
**Other Properties Related to Input:** Context length up to 262K
## Output: **Output Type(s):** Text
**Output Format:** String
## Software Integration: **Supported Runtime Engine(s):**
* SGLang (with `qwen4_exp` model support)
**Supported Hardware Microarchitecture Compatibility:**
* NVIDIA Blackwell (validated on GB300 and B300)
**Preferred Operating System(s):**
* Linux
## Model Version(s): NVFP4 candidate 1.0, quantized with nvidia-modelopt v0.46.0 (snapshot `87c9f8cf83021957d1a1a575c90c9a4eaaf7ef0c`). ## Calibration Dataset: **Link:** [cnn_dailymail](https://huggingface.co/datasets/abisee/cnn_dailymail) (config 3.0.0, train split)
**Properties:** 128 articles (seed 1234), truncated to 512 tokens; MoE-block input activations captured from live SGLang serving (prefill only), 62,139 rows per layer; activation scales by max calibration over 8 seeded sampled batches per part. Representativeness probe: GSM8K train [0:16] x2.
## Post Training Quantization This model was obtained by quantizing Qwen3.8-Flash-Next to NVFP4, ready for inference with SGLang. **Only the routed experts of the 48 main-model MoE layers are quantized** (fused `gate_up_proj` / `down_proj`; 294,912 quantized tensor entries) to NVFP4 W4A4 (E2M1, group size 16, FP8 E4M3 block scales, FP32 global scales, dynamic NVFP4 activations). Attention, QSA, GDN, mHC, shared experts, routers, embeddings, LM head, vision, and **all 31 MTP tensors remain BF16 and byte-identical to the source**. **The PLE n-gram embedding tables use the FP8-quantized versions from the updated `Qwen/Qwen3.8-Flash-Next-FP8` revision** (128 shards F8_E4M3 + per-table scalar scale, dequantized to BF16 at load time); the remaining PLE weights stay BF16. No KV-cache quantization metadata. Checkpoint size is reduced from 360 GB to 135 GB (~2.7x). ## Usage Serve with SGLang (requires a build with `qwen4_exp` support): ```sh python -m sglang.launch_server \ --model-path RadixArk/Qwen3.8-Flash-Next-NVFP4 \ --tp 2 \ --quantization modelopt_fp4 \ --fp4-gemm-backend flashinfer_cutlass \ --page-size 64 \ --mamba-scheduler-strategy extra_buffer \ --mamba-track-interval 64 \ --chunked-prefill-size 4096 \ --max-running-requests 36 \ --context-length 262144 \ --mem-fraction-static 0.80 \ --allow-auto-truncate \ --port 30000 ``` ## Evaluation Results attributed to this exact checkpoint unless noted otherwise: | Eval | Protocol | BF16 reference¹ | NVFP4 (this checkpoint) | |---|---|---|---| | GSM8K | full 1319, t0.6 / top-p 0.95 / max 8192 | 97.12–97.50² | **97.27** (stop 98.86, err 0) | | AIME26 | 30 problems x 8, t1.0 / max 130k | 100 (240/240) | **98.75 pass@1** (majority@8 100, stop 99.17) ³ | ¹ BF16 reference runs were recorded on an earlier checkpoint revision of the same model line; the two revisions' weight deltas are not established, so treat comparisons as indicative, not exact. ² Range across three independent BF16 runs on the earlier revision. ³ Measured on the previous revision of this checkpoint, whose weights differ from the current revision only in the PLE embedding tables. **Behavioral note:** this quantization preserves single-turn accuracy (GSM8K/AIME in-band); long agentic generations tend to run longer than BF16. ## Integrity Evidence All audits pass on this exact checkpoint: structural audit (294,912 routed + 1,562 unchanged + 31 MTP tensors), scale audit (221,184 finite positive scales, min 2.13e-05, max 448.0), unchanged-content byte-equality audit (1,562 tensors / 118.4 GB), deterministic serving smoke. Raw metrics: `gsm8k_metrics.json`, `aime26_metrics.json`; audit reports and details: `qualification-notes.md`, `validate_checkpoint_report.json`, `validate_scales_report.json`, `audit_unchanged_report.json`. ## Model Limitations: The base model was trained on data that contains toxic language and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.