Instructions to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Use Docker

docker model run hf.co/ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF 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-GGUF"
# 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-GGUF",
		"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-GGUF:Q4_K_M

Ollama
How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with Ollama:
```
ollama run hf.co/ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M
```
Unsloth Desktop

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with Docker Model Runner:
```
docker model run hf.co/ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M
```

Lemonade

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Swift-1.5-GLM-5.3-Flash-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ukisai/Swift-1.5-GLM-5.3-Flash-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Website • Learn more • Safetensors • GGUF • Evaluation • Enterprise licensing

Swift 1.5 GLM-5.3-Flash

GGUF quantizations. Derived directly from Swift 1.5 GLM-5.3-Flash with llama.cpp. Run a chosen quantization tier with a current llama.cpp-compatible runtime such as llama-server.

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.

Benchmark	Score		Mean tokens			Median tokens
Benchmark	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
Reasoning effort	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
Benchmark	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.

GGUF quantizations

File	Size	KLD wikitext @512	KLD wikitext @32k	99% KLD @32k	Top-p @32k
Q8_0	341.5 GB	0.0179	0.0073	0.096	97.06%
Q6_K	281.0 GB	0.0192	0.0092	0.124	96.69%
Q6_K_S	268.2 GB	0.0216	0.0092	0.122	96.75%
Q5_K_M	243.7 GB	0.0312	0.0155	0.212	96.16%
Q5_1 †	240.5 GB	0.0428	0.0241	0.305	94.45%
Q5_K_S	227.9 GB	0.0330	0.0175	0.235	95.76%
Q5_0 †	221.4 GB	0.0477	0.0259	0.359	94.42%
Q4_1	202.6 GB	0.0817	0.0472	0.569	92.47%
Q4_K_M	200.8 GB	0.0562	0.0316	0.414	94.29%
IQ4_NL	200.8 GB	0.0617	0.0366	0.460	93.85%
Q4_K_S	189.3 GB	0.0673	0.0359	0.463	93.79%
Q4_0	184.1 GB	0.0971	0.0694	0.831	91.30%
IQ4_XS	177.7 GB	0.0711	0.0409	0.524	93.60%
MXFP4_MOE †	176.1 GB	0.1315	0.0834	1.151	90.70%
IQ3_M	166.6 GB	0.1326	0.0952	1.253	89.76%
Q3_K_L	162.4 GB	0.1273	0.0815	1.099	90.49%
Q3_K_M	155.0 GB	0.1500	0.0948	1.234	89.98%
Q3_K_S	147.4 GB	0.1644	0.1014	1.218	89.47%
IQ3_XS	147.4 GB	0.1728	0.1412	1.851	87.82%
IQ3_S †	140.9 GB	0.2245	0.1947	2.192	85.18%
IQ3_XXS	138.8 GB	0.2182	0.1633	2.054	86.67%
Q2_K	125.7 GB	0.3980	0.2812	3.092	81.98%
IQ2_M	120.6 GB	0.3568	0.2733	3.142	82.76%
Q2_K_S †	110.4 GB	0.6614	0.4992	4.720	75.51%
IQ2_S	107.5 GB	0.4756	0.4783	5.159	77.71%
IQ2_XS	101.4 GB	0.5193	0.5401	5.294	75.84%
IQ2_XXS	96.6 GB	0.5911	0.4556	4.598	76.82%
IQ1_M	83.0 GB	0.8868	0.7291	6.303	70.18%
IQ1_S	74.8 GB	1.0220	1.0424	7.782	64.40%

Mean KL divergence against the Swift 1.5 BF16 source, lower is better. wikitext @512 is wikitext-2 test, 100 windows of 512 tokens. wikitext @32k is wikitext-2 train, 16 windows of 32,768 tokens, scoring the last 512 tokens of each window, so every scored token sees at least 32k tokens of context. 99% KLD is the 99th-percentile divergence on the same 32k run. Top-p is top-token agreement with BF16 on the 32k run. † = llama.cpp's own mix.

Each tier is in its own folder, split into several parts. llama.cpp loads every part from the first one. The vision encoder is in the separate mmproj file:

llama-server -hf ukisai/Swift-1.5-GLM-5.3-Flash-GGUF:Q4_K_M --jinja
# or, after downloading a folder and the mmproj file:
llama-server -m Q4_K_M/Swift-1.5-GLM-5.3-Flash-Q4_K_M-00001-of-00005.gguf \
  --mmproj mmproj-Swift-1.5-GLM-5.3-Flash-BF16.gguf --jinja

Use case	Pick
Maximum fidelity	Q8_0
Long agentic runs, strict tool-call formatting	Q6_K or Q6_K_S
Everyday use, about 200 GB of GPU plus system memory	Q4_K_M
About 180 GB	IQ4_XS
About 160 GB	Q3_K_L

Recipe

All 29 tiers were built with llama.cpp commit 2a66703 from a BF16 conversion of the published Swift 1.5 safetensors. They use the importance matrix and per-tensor type layouts from bartowski/GLM-5.3-Flash-BF16-GGUF; the † tiers use llama.cpp's own mix. Every file was checked against BF16 on the harness above.

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: 3,830

GGUF

Model size

321B params

Architecture

glm5-next

Hardware compatibility

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for ukisai/Swift-1.5-GLM-5.3-Flash-GGUF

Base model

zai-org/GLM-5.3-Flash

Finetuned

ukisai/Swift1.5-GLM-5.3-Flash

Quantized

(3)

this model

ukisai
/

Swift-1.5-GLM-5.3-Flash-GGUF

Swift 1.5 GLM-5.3-Flash

Training approach

Evaluation

Efficiency across reasoning efforts

Quantized models

GGUF quantizations

License and access

Citation

Acknowledgements

Model tree for ukisai/Swift-1.5-GLM-5.3-Flash-GGUF

Dataset used to train ukisai/Swift-1.5-GLM-5.3-Flash-GGUF