Total runs:
177.7K
Run Growth:
-42.6K
Growth Rate:
-22.81%
Sponsored by
AdsCreator.com
inclusionAI / Ling-3.1-flash
Model's Last Updated: October 15 2026
text-generation
Introduction of Ling-3.1-flash
Model Details of Ling-3.1-flash
🤗 Hugging Face | 🤖 ModelScope | 🐙 OpenRouter
Introduction
We're introducing Ling-3.1-flash , an upgraded native hybrid reasoning model designed for complex, real-world tasks. With 560B total parameters, 25B active parameters per token, and a 1M-token context window , Ling-3.1-flash builds upon Ling-3.0-flash with improvements in architecture and performance on real-world tasks.
With stronger real-world capabilities, Ling-3.1-flash shows improvements across a range of real-world tasks, including general-purpose agents, search, office productivity, and software engineering. It demonstrates strong performance on benchmarks such as GDPVal-AA V2.1, HealthBench Professional, FrontierSWE, and Terminal-Bench 4.0 , covering office tasks, healthcare, software engineering, and terminal-based tool use. The model also shows improvements in specialized domains such as finance and materials science.
Model Overview
The table and diagram below summarize the model’s specifications and architecture.
| Architecture | Hybrid-linear MoE |
|---|---|
| Parameter Scale | Total 560B, Activated 25B |
| Transformer Layers | 63 KDA + 9 Gated MLA (7:1) |
| Number of Dense Layers | 4 |
| Number of Routed Experts | 512 |
| Number of Shared Experts | 1 |
| Number of Activated Experts | 8 |
| Attention Heads | 64 |
| Hidden Size | 4096 |
| Expert Intermediate Size | 1280 |
| Dense Intermediate Size | 10240 |
| Vocabulary Size | 157184 |
Evaluation
We have conducted a comprehensive evaluation of Ling-3.1-flash across multiple authoritative benchmarks. Ling-3.1-flash performs strongly on representative coding/agentic benchmarks including Terminal-Bench 4.0 , SWE-bench Pro , FrontierSWE , GDPVal-AA V2.1 , Automationbench , MCP-Atlas , and SkillsBench , among others. In practice, Ling-3.1-flash delivers a strong user experience across Claude Code , Kilo Code , Qwen Code , Hermes Agent , OpenClaw , and other frameworks. Beyond agentic tasks, Ling-3.1-flash performs strongly on real-world tasks across STEM , healthcare , and finance .
We also introduce LingCode , a harness built on the open-source coding agent Pi and enhanced to support complex, long-horizon coding tasks that run for hours, involve hundreds of rounds of tool calls, and require repeated context compaction. LingCode delivers significant performance gains on some coding agent tasks, and we are considering open-sourcing it in the future.
Quickstart
SGLang (Recommended)
The day-0 SGLang image includes Ling-3.1-flash support, MTP (NEXTN) speculative decoding, and Hopper decode optimizations:
docker pull lmsysorg/sglang:dev-ling-3.1-flash
Run Inference
Server (8 × H200, TP8 + EP8):
docker run --gpus all --network host --ipc=host --rm \
lmsysorg/sglang:dev-ling-3.1-flash \
python3 -m sglang.launch_server \
--model-path inclusionAI/Ling-3.1-flash \
--tp 8 --ep 8 \
--speculative-algorithm NEXTN \
--host 0.0.0.0 --port 30000
On 4 × GB300, use
--tp 4 --ep 4
instead. Hopper decode optimizations (optional, ~10% faster batch-1 decode) can be enabled with
SGLANG_ENABLE_FP8_BLOCK_GEMV=1
and
SGLANG_ENABLE_FUSED_KDA_OUTPUT_QUANT=1
.
Client
curl -s http://localhost:30000/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model": "Ling-3.1-flash",
"messages": [{"role": "user", "content": "What is the capital of France?"}],
"chat_template_kwargs": {"enable_thinking": true},
"stream": true,
"temperature": 1.0,
"top_k": 20,
"top_p": 0.95}'
Verified recipes for H200, GB300, H800, H100, and B200, plus HiCache (Mooncake L3), reasoning on/off, tool calling, and measured accuracy/speed numbers, are in the SGLang Ling-3.1-flash cookbook .
vLLM
Environment Preparation
pip install uv
uv venv ~/my_ling_env
source ~/my_ling_env/bin/activate
git clone -b vllm https://github.com/vllm-project/vllm.git
cd vllm
VLLM_USE_PRECOMPILED=1 uv pip install --editable . --torch-backend=auto
Run Inference
Server
vllm serve "$MODEL_PATH" \
--port "$PORT" \
--trust-remote-code \
--served-model-name auto \
--tensor-parallel-size 8 \
--gpu-memory-utilization 0.85 \
--enable-prefix-caching \
--enable-expert-parallel \
--mamba-cache-mode align \
--enable-auto-tool-choice \
--tool-call-parser ling3 \
--reasoning-parser ling3 \
--speculative-config '{"method":"mtp","num_speculative_tokens":3}'
Client
It is recommended to use the sampling parameters (
temperature=1.0
,
top_p=0.95
,
top_k=20
) and set
enable_thinking
to
true
for better performance.
curl -s http://${MASTER_IP}:${PORT}/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model": "auto",
"messages": [{"role": "user", "content": "What is the capital of France?"}],
"chat_template_kwargs": {"enable_thinking": true},
"stream": true,
"temperature": 1.0,
"top_k": 20,
"top_p": 0.95}'
Training Content Summary
The public training-content summary identifying Ling-3.1-flash is available below. Please refer to the document for its covered model versions, training-content scope, summary version, and update date.
This summary concerns training-content disclosure; it does not replace the model’s technical documentation, usage terms, or license.
Runs of inclusionAI Ling-3.1-flash on huggingface.co
1.9K
Total runs
0
24-hour runs
0
3-day runs
0
7-day runs
0
30-day runs
More Information About Ling-3.1-flash huggingface.co Model
Url of Ling-3.1-flash
Ling-3.1-flash huggingface.co Url
Provider of Ling-3.1-flash huggingface.co
Other API from inclusionAI
Updated:April 13 2026
Total runs:
87.7K
Run Growth:
-21.8K
Growth Rate:
-23.19%
Total runs:
23.9K
Run Growth:
-5.3K
Growth Rate:
-22.05%
Total runs:
22.2K
Run Growth:
3.8K
Growth Rate:
16.87%
Total runs:
15.5K
Run Growth:
10.3K
Growth Rate:
64.34%
Total runs:
15.0K
Run Growth:
7.7K
Growth Rate:
51.88%
Total runs:
11.7K
Run Growth:
-9.1K
Growth Rate:
-67.50%
Total runs:
11.3K
Run Growth:
-31.9K
Growth Rate:
-291.50%
Total runs:
10.3K
Run Growth:
-6.4K
Growth Rate:
-64.14%
Total runs:
9.6K
Run Growth:
3.9K
Growth Rate:
40.19%
Total runs:
8.8K
Run Growth:
6.5K
Growth Rate:
74.04%
Total runs:
5.4K
Run Growth:
4.3K
Growth Rate:
84.88%
Total runs:
5.4K
Run Growth:
-6.1K
Growth Rate:
-108.06%
Total runs:
5.3K
Run Growth:
917
Growth Rate:
17.93%
Total runs:
3.8K
Run Growth:
2.6K
Growth Rate:
65.44%
Total runs:
3.2K
Run Growth:
263
Growth Rate:
8.91%
Total runs:
3.2K
Run Growth:
-581
Growth Rate:
-17.72%
Total runs:
3.0K
Run Growth:
-1.2K
Growth Rate:
-40.26%
Total runs:
2.6K
Run Growth:
2.5K
Growth Rate:
97.66%
Total runs:
2.5K
Run Growth:
-12.3K
Growth Rate:
-468.64%
Total runs:
2.3K
Run Growth:
-441
Growth Rate:
-19.39%
Total runs:
2.3K
Run Growth:
1.5K
Growth Rate:
63.01%
Total runs:
2.2K
Run Growth:
1.8K
Growth Rate:
82.35%
Total runs:
2.2K
Run Growth:
-923
Growth Rate:
-42.20%
Total runs:
2.2K
Run Growth:
963
Growth Rate:
51.06%
Total runs:
2.0K
Run Growth:
479
Growth Rate:
23.46%
Total runs:
2.0K
Run Growth:
-196
Growth Rate:
-9.80%
Total runs:
1.9K
Run Growth:
1.8K
Growth Rate:
91.97%
Total runs:
1.8K
Run Growth:
1.2K
Growth Rate:
66.29%
Total runs:
1.7K
Run Growth:
-6.8K
Growth Rate:
-385.84%
Total runs:
1.7K
Run Growth:
-8.9K
Growth Rate:
-515.15%
Total runs:
1.7K
Run Growth:
298
Growth Rate:
17.80%
Total runs:
1.6K
Run Growth:
-2.1K
Growth Rate:
-135.35%
Total runs:
1.5K
Run Growth:
902
Growth Rate:
62.25%
Total runs:
1.4K
Run Growth:
-300
Growth Rate:
-21.52%
Total runs:
1.4K
Run Growth:
321
Growth Rate:
23.90%
Total runs:
1.3K
Run Growth:
-5.6K
Growth Rate:
-427.00%
Total runs:
1.3K
Run Growth:
398
Growth Rate:
31.17%
Total runs:
1.3K
Run Growth:
-1.5K
Growth Rate:
-123.12%
Total runs:
1.2K
Run Growth:
668
Growth Rate:
53.91%
Total runs:
1.1K
Run Growth:
-516
Growth Rate:
-45.62%
Total runs:
1.0K
Run Growth:
470
Growth Rate:
100.00%
Total runs:
991
Run Growth:
912
Growth Rate:
92.03%
Total runs:
859
Run Growth:
802
Growth Rate:
93.36%
Total runs:
828
Run Growth:
820
Growth Rate:
99.03%
Total runs:
787
Run Growth:
-18
Growth Rate:
-2.30%
Total runs:
759
Run Growth:
298
Growth Rate:
39.26%
Total runs:
741
Run Growth:
186
Growth Rate:
25.41%
Total runs:
700
Run Growth:
379
Growth Rate:
54.85%
Total runs:
677
Run Growth:
-1.1K
Growth Rate:
-165.24%
Total runs:
676
Run Growth:
74
Growth Rate:
11.37%
Total runs:
622
Run Growth:
305
Growth Rate:
49.51%
Total runs:
597
Run Growth:
529
Growth Rate:
88.61%
Total runs:
560
Run Growth:
-897
Growth Rate:
-157.92%
Total runs:
552
Run Growth:
-461
Growth Rate:
-84.43%
Total runs:
540
Run Growth:
115
Growth Rate:
20.03%
Total runs:
475
Run Growth:
462
Growth Rate:
100.00%
Total runs:
468
Run Growth:
265
Growth Rate:
56.62%
Total runs:
464
Run Growth:
-319
Growth Rate:
-65.91%
Total runs:
434
Run Growth:
-1.7K
Growth Rate:
-385.55%
Total runs:
432
Run Growth:
373
Growth Rate:
86.34%
Total runs:
426
Run Growth:
-2.0K
Growth Rate:
-471.19%
Total runs:
398
Run Growth:
-1.5K
Growth Rate:
-372.59%
Total runs:
386
Run Growth:
133
Growth Rate:
35.37%
Total runs:
380
Run Growth:
255
Growth Rate:
67.46%
Total runs:
377
Run Growth:
-1.9K
Growth Rate:
-512.64%
Total runs:
325
Run Growth:
-1.8K
Growth Rate:
-563.35%
Total runs:
325
Run Growth:
-215
Growth Rate:
-67.40%
Total runs:
325
Run Growth:
-2.3K
Growth Rate:
-735.74%
Total runs:
324
Run Growth:
-160
Growth Rate:
-48.93%
Total runs:
318
Run Growth:
236
Growth Rate:
74.21%
Total runs:
315
Run Growth:
-1.3K
Growth Rate:
-405.10%
Total runs:
314
Run Growth:
45
Growth Rate:
14.71%
Total runs:
308
Run Growth:
144
Growth Rate:
47.21%
Total runs:
302
Run Growth:
302
Growth Rate:
100.00%
Total runs:
294
Run Growth:
-558
Growth Rate:
-190.44%
Total runs:
292
Run Growth:
181
Growth Rate:
46.53%
Total runs:
285
Run Growth:
-1.7K
Growth Rate:
-601.06%
Total runs:
285
Run Growth:
134
Growth Rate:
52.55%
Total runs:
285
Run Growth:
-27
Growth Rate:
-9.38%
Total runs:
285
Run Growth:
268
Growth Rate:
94.04%
Total runs:
281
Run Growth:
254
Growth Rate:
90.39%
Total runs:
279
Run Growth:
175
Growth Rate:
62.50%
Total runs:
270
Run Growth:
39
Growth Rate:
14.50%
Total runs:
269
Run Growth:
153
Growth Rate:
49.04%
Total runs:
267
Run Growth:
-118
Growth Rate:
-43.54%
Total runs:
264
Run Growth:
133
Growth Rate:
52.57%
Total runs:
251
Run Growth:
-286
Growth Rate:
-112.60%