Total runs:
14.0K
Run Growth:
14.0K
Growth Rate:
99.81%
Sponsored by
iMean AI Trip Planner
Ex0bit / Step-3.5-Flash-PRISM-PRO
Model's Last Updated: February 10 2026
text-generation
Introduction of Step-3.5-Flash-PRISM-PRO
Model Details of Step-3.5-Flash-PRISM-PRO
-blue){kind=icon}
Step-3.5-Flash-PRISM-PRO
Full Production unrestricted/unchained PRISM-PRO version of StepFun's Step 3.5 Flash intended particularly for full over-refusal and propaganda mechanisms removal with the SOTA PRISM-PRO pipeline.
For Full Custom Production PRISM versions & tensors reach out.
Free PRISM-Lite version avialable here: https://hf.co/Ex0bit/Step-3.5-Flash-PRISM
☕ Support Our Work
If you enjoy our work and find it useful, please consider sponsoring or supporting us!
| Option | Description |
|---|---|
| PRISM VIP Membership | Access to all PRISM models |
| Bitcoin |
bc1qarq2pyn4psjpcxzp2ghgwaq6y2h4e53q232x8r
|
Model Highlights
- PRISM Ablation — State-of-the-art technique that removes over-refusal behaviors while preserving model capabilities
- 196B MoE Architecture — 196 billion total parameters with only 11 billion active per token across 288 fine-grained routed experts + 1 shared expert
- Multi-Token Prediction (MTP-3) — Predicts 4 tokens simultaneously, achieving 100–300 tok/s typical throughput (peaking at 350 tok/s)
- 256K Context Window — Cost-efficient long context via 3:1 Sliding Window Attention (SWA) ratio
- Frontier Reasoning & Coding — 97.3 on AIME 2025, 74.4% on SWE-bench Verified, 51.0% on Terminal-Bench 2.0
- Accessible Local Deployment — Runs on high-end consumer hardware (Mac Studio M4 Max, NVIDIA DGX Spark)
Model Architecture
| Specification | Value |
|---|---|
| Architecture | Sparse Mixture-of-Experts (MoE) |
| Backbone | 45-layer Transformer (4,096 hidden dim) |
| Total Parameters | 196.81B (196B Backbone + 0.81B Head) |
| Activated Parameters | ~11B (per token) |
| Routed Experts per Layer | 288 |
| Shared Experts | 1 (always active) |
| Selected Experts per Token | Top-8 |
| Vocabulary Size | 128,896 |
| Context Length | 256K |
| Attention | Hybrid SWA (3:1 SWA-to-Full ratio) |
| MTP Head | Sliding-window attention + dense FFN (4 tokens/pass) |
Benchmarks
| Benchmark | Step 3.5 Flash | DeepSeek V3.2 | Kimi K2.5 | GLM-4.7 | MiniMax M2.1 |
|---|---|---|---|---|---|
| Agent | |||||
| τ²-Bench | 88.2 | 80.3 | 85.4 | 87.4 | 86.6 |
| BrowseComp | 51.6 | 51.4 | 60.6 | 52.0 | 47.4 |
| GAIA (no file) | 84.5 | 75.1 | 75.9 | 61.9 | 64.3 |
| xbench-DeepSearch (2025.05) | 83.7 | 78.0 | 76.7 | 72.0 | 68.7 |
| Reasoning | |||||
| AIME 2025 | 97.3 | 93.1 | 96.1 | 95.7 | 83.0 |
| HMMT 2025 (Feb.) | 98.4 | 92.5 | 95.4 | 97.1 | 71.0 |
| IMOAnswerBench | 85.4 | 78.3 | 81.8 | 82.0 | 60.4 |
| Coding | |||||
| LiveCodeBench-V6 | 86.4 | 83.3 | 85.0 | 84.9 | — |
| SWE-bench Verified | 74.4 | 73.1 | 76.8 | 73.8 | 74.0 |
| Terminal-Bench 2.0 | 51.0 | 46.4 | 50.8 | 41.0 | 47.9 |
llama.cpp (GGUF)
For local deployment (requires ~120 GB VRAM for int4, smaller quants are available):
./llama-cli -m Step-3.5-Flash-PRISM-PRO-IQ4_NL.gguf --jinja
Recommended Parameters
| Use Case | Temperature | Top-P | Max New Tokens |
|---|---|---|---|
| Agentic / Reasoning / Coding | 1.0 | 0.95 | 32768 |
| General Chat | 0.6 | 0.95 | 4096 |
Hardware Requirements
| Setup | Details |
|---|---|
| Safetensors (Full) | 8x H100/A100 80GB with tensor parallelism |
| BF16 (Full) | 8x H100/A100 80GB with tensor parallelism |
| FP8 Quantized | 8x A100 80GB with expert parallelism |
| GGUF INT4 (Local) | ~120 GB unified memory (Mac Studio M4 Max 128GB, DGX Spark, AMD Ryzen AI Max+ 395) |
License
This model is released under the PRISM Research License .
Acknowledgments
Based on Step 3.5 Flash by StepFun AI . See the technical report and blog post for more details on the base model.
Runs of Ex0bit Step-3.5-Flash-PRISM-PRO on huggingface.co
30
Total runs
0
24-hour runs
2
3-day runs
5
7-day runs
18
30-day runs
More Information About Step-3.5-Flash-PRISM-PRO huggingface.co Model
Url of Step-3.5-Flash-PRISM-PRO
Step-3.5-Flash-PRISM-PRO huggingface.co Url
Provider of Step-3.5-Flash-PRISM-PRO huggingface.co
Other API from Ex0bit
Updated:February 14 2026
Total runs:
1.8K
Run Growth:
1.8K
Growth Rate:
100.00%
Total runs:
1.4K
Run Growth:
-1.0K
Growth Rate:
-73.57%
Total runs:
1.3K
Run Growth:
-17.4K
Growth Rate:
-1360.52%
Total runs:
1.2K
Run Growth:
1.1K
Growth Rate:
92.35%
Total runs:
1.2K
Run Growth:
578
Growth Rate:
48.49%
Total runs:
888
Run Growth:
403
Growth Rate:
45.38%
Total runs:
565
Run Growth:
-272
Growth Rate:
-48.14%
Total runs:
556
Run Growth:
121
Growth Rate:
21.45%
Total runs:
528
Run Growth:
167
Growth Rate:
31.63%
Total runs:
518
Run Growth:
-245
Growth Rate:
-47.30%
Total runs:
513
Run Growth:
223
Growth Rate:
43.47%
Total runs:
412
Run Growth:
-3.1K
Growth Rate:
-754.13%
Total runs:
294
Run Growth:
19
Growth Rate:
6.46%
Total runs:
247
Run Growth:
16
Growth Rate:
6.48%
Total runs:
227
Run Growth:
112
Growth Rate:
47.66%
Total runs:
212
Run Growth:
10
Growth Rate:
4.72%
Total runs:
113
Run Growth:
105
Growth Rate:
92.92%
Total runs:
87
Run Growth:
24
Growth Rate:
27.59%
Total runs:
48
Run Growth:
-26
Growth Rate:
-54.17%
Total runs:
44
Run Growth:
-26
Growth Rate:
-59.09%
Total runs:
39
Run Growth:
13
Growth Rate:
33.33%
Total runs:
31
Run Growth:
-119
Growth Rate:
-383.87%
Total runs:
23
Run Growth:
21
Growth Rate:
91.30%
Total runs:
18
Run Growth:
-2.5K
Growth Rate:
-13922.22%
Total runs:
16
Run Growth:
6
Growth Rate:
37.50%
Total runs:
10
Run Growth:
-23
Growth Rate:
-230.00%