Rio 3.0 Open Mini
Rio 3.0 Open Mini is a frontier-class reasoning model developed by
IplanRIO, the municipal IT company of Rio de Janeiro's city government. Built through distillation on top of Qwen3-4B-Thinking-2507 using reasoning traces from our to be announced Rio 3.0 model, Rio 3.0 Open achieves state-of-the-art results across mathematics, STEM, and code benchmarks — surpassing its base model by significant margins and competing with models far larger than itself.
Rio 3.0 Open Mini features
SwiReasoning, a training-free inference framework based on
Shi et al. (2025) that dynamically switches between explicit chain-of-thought and latent-space reasoning, guided by entropy-based confidence signals. This enables both higher accuracy and dramatically improved token efficiency. This model was explicitly trained to maximize the efficiency gained via latent reasoning.
Key Features
- 4B total parameters
- 262,144 token context window
- SwiReasoning integration — dynamic explicit/latent reasoning switching for Pareto-superior accuracy and efficiency
- Distilled from Qwen3-4B-Thinking-2507 with traces from Rio 3.0
- Multilingual — strong performance in Portuguese, English, Chinese, and dozens of other languages
- MIT License — fully open for commercial and research use
Benchmark Results
Mathematics & STEM
| Model | GPQA Diamond | LiveCodeBench | Composite Math* | AIME 2025 | AIME 2026 I | HMMT 2025 I | HMMT 2025 II | BRUMO 2025 | CMIMC 2025 | SMT 2025 |
|---|
| Rio 3.0 Open Mini | 71.90% | 63.50% | 78.11% | 89.17% | 75.00% | 73.33% | 79.17% | 85.83% | 66.88% | 77.36% |
| Rio 3.0 Open Mini (w/o latent) | 70.10% | 62.00% | 75.53% | 85.83% | 75.83% | 66.67% | 74.17% | 84.17% | 63.75% | 78.30% |
| Qwen3-4B-2507 (base) | 65.80% | 55.20% | 71.12% | 81.67% | 70.83% | 55.83% | 73.33% | 81.67% | 60.00% | 74.53% |
| Qwen3-30B-A3B-2507 | 73.40% | 66.00% | 76.08% | 82.50% | 76.67% | 70.83% | 75.83% | 85.00% | 66.25% | 75.47% |
| GPT OSS 20B | 71.50% | 70.26% | 82.34% | 89.17% | 85.00% | 76.67% | 83.33% | 86.67% | 72.50% | 83.02% |
*Composite Math is the average across all other mathematics benchmarks in this table.
Rio Model Family Comparison
| Model | GPQA Diamond | LiveCodeBench | Composite Math* | AIME 2025 |
|---|
| Rio 3.0 Open | 85.10% | 76.00% | 91.78% | 96.67% |
| Rio 2.5 Open | 77.20% | 69.60% | 87.53% | 93.33% |
| Rio 3.0 Open Mini | 71.90% | 63.50% | 78.11% | 89.17% |
Gains Over Base Model (Qwen3-4B-2507)
| Benchmark | Base Model | Rio 3.0 Open Mini | Δ |
|---|
| GPQA Diamond | 65.80% | 71.90% | +6.10% |
| LiveCodeBench | 55.20% | 63.50% | +8.30% |
| Composite Math | 71.12% | 78.11% | +6.99% |
| AIME 2025 | 81.67% | 89.17% | +7.50% |
| AIME 2026 I | 70.83% | 75.00% | +4.17% |
| HMMT 2025 I | 55.83% | 73.33% | +17.50% |
| BRUMO 2025 | 81.67% | 85.83% | +4.16% |
| CMIMC 2025 | 60.00% | 66.88% | +6.88% |
| SMT 2025 | 74.53% | 77.36% | +2.83% |
SwiReasoning: Latent/Explicit Reasoning
Rio 3.0 Open Mini integrates
SwiReasoning (Shi et al., 2025), a training-free inference framework that dynamically alternates between two reasoning modes:
- Explicit reasoning — standard chain-of-thought in natural language, where the model commits tokens to a single reasoning path
- Latent reasoning — continuous reasoning in hidden space, where the model explores multiple implicit paths simultaneously without emitting tokens
The switching is governed by block-wise confidence estimated from entropy trends in the next-token distribution. When confidence is low (entropy trending upward), the model enters latent mode to explore alternatives. When confidence recovers, it switches back to explicit mode to commit to a solution.
This approach achieves a Pareto-superior trade-off: higher accuracy at unlimited budgets and dramatically better token efficiency under constrained budgets.
The benchmark table above includes (w/o latent) rows showing performance with standard explicit-only reasoning, demonstrating the consistent gains from SwiReasoning across all benchmarks.
How to Use
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model_name = "prefeitura-rio/Rio-3.0-Open-Mini"
4
5tokenizer = AutoTokenizer.from_pretrained(model_name)
6model = AutoModelForCausalLM.from_pretrained(
7 model_name,
8 torch_dtype="auto",
9 device_map="auto"
10)
11
12prompt = "Write a poem about Rio de Janeiro."
13
14messages = [
15 {"role": "user", "content": prompt}
16]
17
18text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
19inputs = tokenizer([text], return_tensors="pt").to(model.device)
20
21outputs = model.generate(
22 **inputs,
23 max_new_tokens=81920,
24 temperature=0.6,
25 top_p=0.95,
26)
27
28response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True)
29print(response)
Using with vLLM
1vllm serve prefeitura-rio/Rio-3.0-Open-Mini \
2 --tensor-parallel-size 4 \
3 --max-model-len 262144 \
4 --trust-remote-code
Using with SGLang
1python -m sglang.launch_server \
2 --model-path prefeitura-rio/Rio-3.0-Open-Mini \
3 --tp 4 \
4 --context-length 262144 \
5 --trust-remote-code
Model Details
| |
|---|
| Developer | IplanRIO — Empresa Municipal de Informática e Planejamento S.A. |
| Base Model | Qwen3-4B-Thinking-2507 |
| Architecture | Transformer |
| Total Parameters | ~4B |
| Context Length | 262,144 tokens |
| Default Max Output Length | 81,920 tokens |
| Training Method | Distillation |
| Inference Enhancement | SwiReasoning (latent/explicit switching) |
| License | MIT |
| Languages | Multilingual (en, pt, zh, ja, ko, fr, de, es, ar, and more) |
Citation
If you use SwiReasoning, please also cite:
1@misc{shi2025swireasoning,
2 title={SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs},
3 author={Dachuan Shi et al.},
4 year={2025},
5 eprint={2510.05069},
6 archivePrefix={arXiv},
7 primaryClass={cs.CL}
8}
Acknowledgments
Rio 3.0 Open Mini is built upon the exceptional work of the
Qwen Team and their Qwen3 model family. We also acknowledge the authors of
SwiReasoning for their innovative inference framework.
Developed in Rio de Janeiro 🇧🇷 by
IplanRIO.
