Instructions to use ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4") model = AutoModelForMultimodalLM.from_pretrained("ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4
- SGLang
How to use ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 with Docker Model Runner:
docker model run hf.co/ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4
Swift 1.5 GLM-5.3-Flash
NVFP4. Quantized from Swift 1.5 GLM-5.3-Flash with NVIDIA Model Optimizer, following NVIDIA's published recipe: NVFP4 on the routed experts and dense MLP, with FP8 KV cache. Native NVFP4 execution requires compatible NVIDIA Blackwell hardware.
Swift 1.5 GLM-5.3-Flash is UkisAI's reasoning-efficient derivative of GLM-5.3-Flash. It uses 57.7% fewer thinking tokens, with a 2.4x speed up while keeping the accuracy loss <1% vs base on max reasoning effort.
The prompt is a sample from LiveCodeBench v6
Training approach
We made Swift GLM-5.3-Flash efficient by figuring out which tokens were linked to pathological overthinking and penalizing them without "attacking" the reasoning length directly, then regained the accuracy with further post-training, leading to "compressed" token usage while maintaining accuracy.
Swift 1.5 produces shorter reasoning traces. In our testing, we also observe fewer overthinking errors.
This release also features our previously mentioned post-training methods adapted specifically for coding and long-horizon agent work such as terminal use and software engineering.
Our training data is viewable here: https://huggingface.co/datasets/ukisai/Qwen3.8-27B-multi-turn-agent-sft albeit it is not used out of the box, but rather re-sampled, turned into proper RL environments etc.
Evaluation
The external results below compare GLM-5.3-Flash, the foundation base model, and Swift 1.5. Both models use the same saved evaluation protocols, and all scores are reported as final aggregate percentages.
The main table reports BF16 results. Quantized results are in Quantized models.
| Benchmark | Score | Mean tokens | Median tokens | |||
|---|---|---|---|---|---|---|
| Base | Swift 1.5 | Base | Swift 1.5 | Reduction | Reduction | |
| General reasoning | ||||||
| GPQA-Diamond | 90.51% | 89.90% | 10,248 | 6,052 | ↓ 40.9% | ↓ 57.7% |
| MMLU-Pro | 87.38% | 86.25% | 2,755 | 1,064 | ↓ 61.4% | ↓ 44.4% |
| C-Eval | 93.42% | 93.68% | 1,214 | 752 | ↓ 38.0% | ↓ 19.9% |
| IFBench | 56.40% | 55.73% | 11,660 | 8,299 | ↓ 28.8% | ↓ 22.9% |
| Mathematics | ||||||
| AIME 2026 | 93.33% | 92.67% | 24,193 | 18,145 | ↓ 25.0% | ↓ 45.8% |
| HMMT (Nov 2025) | 95.33% | 91.33% | 25,924 | 18,797 | ↓ 27.5% | ↓ 31.7% |
| Coding | ||||||
| LiveCodeBench v6 | 86.69% | 87.94% | 21,506 | 14,645 | ↓ 31.9% | ↓ 36.1% |
| Agentic coding | ||||||
| Terminal-Bench 2.1* | 76.40% | 75.96% | 118,878 | 79,473 | ↓ 33.1% | ↓ 36.8% |
| DeepSWE v1.1 | 46.02% | 48.67% | 75,671 | 59,513 | ↓ 21.4% | ↓ 20.9% |
* Note: Terminal-Bench 2.1 ran with a 5-hour limit (18,000 seconds) on every task, not the native per-task limits, for both base and Swift 1.5.
How to reproduce
Serving: BF16 · vLLM · GLM-4.7 reasoning parser · context 262,144 · reasoning effort max · MTP disabled.
Sampling: temperature 1.0 · top_p 0.95 · presence_penalty 0 · repetition_penalty 1.
Benchmarks: five seeds (0–4) for the question benchmarks. LiveCodeBench is full release v6 mean pass@1 over seeds.
Terminal-Bench 2.1: Harbor 0.20.0 / Terminus-2, pinned 89-task dataset, five attempts per task, temperature 1, top_p 1, 18,000-second task limit.
DeepSWE v1.1: 113 tasks × 3 attempts, Pier 0.3.1 with mini-swe-agent 2.4.6, GLM-4.7 tool-call parser.
| Benchmark | Output cap |
|---|---|
| GPQA-Diamond | 100,000 |
| MMLU-Pro | 100,000 |
| C-Eval | 16,384 |
| IFBench | 81,920 |
| AIME 2026 | 250,000 |
| HMMT Nov 2025 | 250,000 |
| LiveCodeBench v6 | 100,000 |
Efficiency across reasoning efforts
GPQA-Diamond at each reasoning_effort setting, Swift 1.5 against the base at the same setting:
| Reasoning effort | Score | Mean tokens | Median tokens | |||
|---|---|---|---|---|---|---|
| Base | Swift 1.5 | Base | Swift 1.5 | Reduction | Reduction | |
| Max | 90.51% | 89.90% | 10,248 | 6,052 | ↓ 40.9% | ↓ 57.7% |
| High | 88.79% | 86.97% | 7,623 | 4,158 | ↓ 45.5% | ↓ 31.1% |
| Low | 77.27% | 73.33% | 2,815 | 1,377 | ↓ 51.1% | ↓ 13.7% |
At max, Swift 1.5 trails base by 0.61 percentage points while using 40.9% fewer mean and 57.7% fewer median thinking tokens. High and low save tokens but also lose 1.82 and 3.94 percentage points respectively.
Quantized models
| Format | Repository | Runtime |
|---|---|---|
| NVFP4 | Swift-1.5-GLM-5.3-Flash-NVFP4 | NVIDIA Blackwell |
| AWQ INT4 (W4A16) | Swift-1.5-GLM-5.3-Flash-W4A16-AWQ | vLLM (compressed-tensors) |
| GGUF | Swift-1.5-GLM-5.3-Flash-GGUF | llama.cpp |
Each quantized Swift 1.5 against a quantized base of the same format, on the same protocol as the BF16 evaluation:
| Benchmark | Score | Mean tokens | Median tokens | |||
|---|---|---|---|---|---|---|
| Base | Swift 1.5 | Base | Swift 1.5 | Reduction | Reduction | |
| NVFP4 | ||||||
| GPQA-Diamond | 90.10% | 90.20% | 13,069 | 7,148 | ↓ 45.3% | ↓ 59.5% |
| AIME 2026 | 94.00% | 91.33% | 30,855 | 29,768 | ↓ 3.5% | ↓ 48.3% |
| W4A16-AWQ | ||||||
| GPQA-Diamond | 90.81% | 89.19% | 10,601 | 6,718 | ↓ 36.6% | ↓ 55.3% |
| AIME 2026 | 92.67% | 94.00% | 20,138 | 22,546 | ↑ 12.0% | ↓ 46.8% |
NVFP4 base: nvidia/GLM-5.3-Flash-NVFP4, the same recipe. W4A16-AWQ base: wtdcode/GLM-5.3-Flash-AWQ-W4A16, the closest public AWQ.
How to use
Native NVFP4 execution requires compatible NVIDIA Blackwell hardware and a runtime with GLM-5.3-Flash ModelOpt support. The quantized evaluation uses FP8 KV cache (--kv-cache-dtype fp8).
For installation and hardware-specific settings, see the vLLM GLM-5.3-Flash recipe and NVIDIA's GLM-5.3-Flash NVFP4 reference.
We evaluated this format on 8x B200 with a GLM-5.3-Flash vLLM build (vllm/vllm-openai:glm53-flash-x86_64-cu129), tensor parallelism 8, a 262,144-token context and MTP disabled. Load the checkpoint directly from this repository:
vllm serve ukisai/Swift-1.5-GLM-5.3-Flash-NVFP4 \
--dtype bfloat16 \
--tensor-parallel-size 8 \
--kv-cache-dtype fp8 \
--max-model-len 262144 \
--reasoning-parser glm47 \
--enable-auto-tool-choice \
--tool-call-parser glm47 \
--port 8000
Reasoning effort is set per request through the chat template, for example "chat_template_kwargs": {"reasoning_effort": "high"}. max is the default; high and low are also supported.
Quantization
Built with NVIDIA Model Optimizer (main @ 90ba9fb) using NVIDIA's published recipe for this architecture,
GLM-5.3-Flash-BF16/ptq/nvfp4_experts_dense_mlp-kv_fp8_cast,
the same recipe as nvidia/GLM-5.3-Flash-NVFP4:
- NVFP4 (W4A4, group size 16): the routed experts (layers 3-44) and the dense MLP (layers 0-2).
- FP8 KV cache (cast mode).
- BF16: attention (KDA and sparse MLA), shared experts, router, vision tower, embeddings,
lm_headand the MTP head. - Calibration: max, 1,024 samples of 512 tokens from cnn_dailymail and Nemotron-Post-Training-Dataset-v2 (512 each).
The one-layer MTP head (layer 45) is included unquantized in BF16, as in NVIDIA's release, so self-speculative decoding works:
--speculative-config '{"method":"mtp","num_speculative_tokens":5}'
License and access
Swift 1.5 GLM-5.3-Flash is a derivative of GLM-5.3-Flash (Copyright (c) 2026 Z.AI Co., Ltd, MIT License). UkisAI's contribution, including the adapted weights, is licensed under the Swift Open License v1.0. See NOTICE for the change notice and attribution details.
Personal, research, educational, evaluation, and commercial use of the Swift contribution are free for individuals and organizations with gross annual revenue, including affiliates, of up to US$1,000,000. Above that threshold, commercial use requires a separate Swift Enterprise License. Contact UkisAI for terms.
The base model's MIT License still applies to the base model. Nothing in the Swift Open License limits your rights in GLM-5.3-Flash itself under the MIT License.
Citation
@misc{swift-1.5-glm-5.3-flash,
title = {Swift 1.5 GLM-5.3-Flash},
author = {UkisAI},
year = {2026},
url = {https://huggingface.co/ukisai/Swift1.5-GLM-5.3-Flash}
}
Acknowledgements
We acknowledge Amazon Web Services for providing compute credits and infrastructure support for Swift's development, training, and evaluation.
- Downloads last month
- 299
