Total runs:
76.8K
Run Growth:
4.6K
Growth Rate:
5.94%
Sponsored by
APIMart
jdopensource / JoyAI-LLM-Flash-Base
Model's Last Updated: February 15 2026
text-generation
Introduction of JoyAI-LLM-Flash-Base
Model Details of JoyAI-LLM-Flash-Base
1. Model Introduction
JoyAI-LLM Flash-Base is a state-of-the-art mixture-of-experts (MoE) language model with 3 billion activated parameters and 48 billion total parameters. Trained with the Muon optimizer, JoyAI Flash-base achieves exceptional performance across frontier knowledge, reasoning, and coding tasks while being meticulously optimized for agentic capabilities. JoyAI-LLM Flash series aim to accelarate high-throughput, latency-sensitive applications where cost per query must remain minimal.
Key Features
- Training-Inference Collaboration: apply Muon optimizer with dense MTP, develop novel optimization techniques to resolve instabilities while scaling up, delivering 1.3× to 1.7× the throughput of the non-MTP version.
- Agentic Intelligence: Specifically designed for tool use, reasoning, and autonomous problem-solving.
2. Model Summary
| Architecture | Mixture-of-Experts (MoE) |
| Total Parameters | 48B |
| Activated Parameters | 3B |
| Number of Layers (Dense layer included) | 40 |
| Number of Dense Layers | 1 |
| Attention Hidden Dimension | 2048 |
| MoE Hidden Dimension (per Expert) | 768 |
| Number of Attention Heads | 32 |
| Number of Experts | 256 |
| Selected Experts per Token | 8 |
| Number of Shared Experts | 1 |
| Vocabulary Size | 129K |
| Context Length | 128K |
| Attention Mechanism | MLA |
| Activation Function | SwiGLU |
3. Evaluation Results
| Benchmark | JoyAI-LLM Flash-base | Qwen3-30B-A3B-base |
|---|---|---|
| MMLU | 84.70 | 82.12 |
| MMLU-Pro | 73.14 | 61.76 |
| CMMLU | 83.09 | 83.60 |
| HumanEval | 85.37 | 87.80 |
| LiveCodeBench | 39.91 | 37.34 |
| GSM8K | 88.78 | 90.37 |
| MATH | 78.16 | 59.60 |
| MATH 500 | 77.00 | 58.00 |
4. License
Both the code repository and the model weights are released under the Modified MIT License .
Runs of jdopensource JoyAI-LLM-Flash-Base on huggingface.co
106
Total runs
0
24-hour runs
0
3-day runs
7
7-day runs
53
30-day runs
More Information About JoyAI-LLM-Flash-Base huggingface.co Model
Url of JoyAI-LLM-Flash-Base
JoyAI-LLM-Flash-Base huggingface.co Url
Provider of JoyAI-LLM-Flash-Base huggingface.co
Other API from jdopensource
Updated:May 09 2026
Total runs:
850
Run Growth:
-72
Growth Rate:
-8.47%
Total runs:
657
Run Growth:
-7.4K
Growth Rate:
-1130.59%
Total runs:
452
Run Growth:
121
Growth Rate:
26.77%
Total runs:
380
Run Growth:
-1.1K
Growth Rate:
-294.74%
Total runs:
286
Run Growth:
94
Growth Rate:
32.87%
Total runs:
225
Run Growth:
-314
Growth Rate:
-139.56%
Total runs:
221
Run Growth:
-11
Growth Rate:
-4.98%
Total runs:
187
Run Growth:
-128
Growth Rate:
-68.45%
Total runs:
121
Run Growth:
-45
Growth Rate:
-37.19%
Total runs:
98
Run Growth:
20
Growth Rate:
20.41%
Total runs:
87
Run Growth:
47
Growth Rate:
54.02%
Total runs:
86
Run Growth:
-1
Growth Rate:
-1.16%
Total runs:
75
Run Growth:
19
Growth Rate:
25.33%
Total runs:
69
Run Growth:
13
Growth Rate:
18.84%
Total runs:
58
Run Growth:
-22
Growth Rate:
-37.93%
Total runs:
41
Run Growth:
1
Growth Rate:
2.44%
Total runs:
30
Run Growth:
6
Growth Rate:
20.00%
Total runs:
28
Run Growth:
-13
Growth Rate:
-46.43%
Total runs:
27
Run Growth:
9
Growth Rate:
33.33%
Total runs:
16
Run Growth:
3
Growth Rate:
18.75%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%