prefeitura-rio

Rio-2.5-Open

Key Features

30B total / 3B active parameters (Mixture-of-Experts)
262,144 token context window
SwiReasoning integration — dynamic explicit/latent reasoning switching for Pareto-superior accuracy and efficiency
Distilled from Qwen3-30B-A3B-Thinking-2507 with traces from Rio 2.5
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

Table with columns: Model, GPQA Diamond, LiveCodeBench, Composite Math*, AIME 2025, AIME 2026 I, HMMT 2025 I, HMMT 2025 II, BRUMO 2025, CMIMC 2025, SMT 2025
Model	GPQA Diamond	LiveCodeBench	Composite Math*	AIME 2025	AIME 2026 I	HMMT 2025 I	HMMT 2025 II	BRUMO 2025	CMIMC 2025	SMT 2025
Rio 2.5 Open	77.20%	69.60%	87.53%	93.33%	89.17%	83.33%	90.83%	88.33%	83.75%	83.96%
Rio 2.5 Open (w/o latent)	75.80%	69.40%	82.17%	90.00%	83.33%	76.67%	84.17%	85.83%	75.00%

*Composite Math is the average across all other mathematics benchmarks in this table.

Rio Model Family Comparison

Table with columns: Model, GPQA Diamond, LiveCodeBench, Composite Math*, AIME 2025
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%

Gains Over Base Model (Qwen3-30B-A3B-Thinking-2507)

Table with columns: Benchmark, Base Model, Rio 2.5 Open, Δ
Benchmark	Base Model	Rio 2.5 Open	Δ
GPQA Diamond	73.40%	77.20%	+3.80%
LiveCodeBench	66.00%	69.60%	+3.60%
Composite Math	76.08%	87.53%	+11.45%
AIME 2025	82.50%	93.33%

SwiReasoning: Latent/Explicit Reasoning

Rio 2.5 Open 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

python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "prefeitura-rio/Rio-2.5-Open"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)

prompt = "Write a poem about Rio de Janeiro."

messages = [
    {"role": "user", "content": prompt}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer([text], return_tensors="pt").to(model.device)

outputs = model.generate(
    **inputs,
    max_new_tokens=81920,
    temperature=0.6,
    top_p=0.95,
)

response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True)
print(response)

Using with vLLM

bash
vllm serve prefeitura-rio/Rio-2.5-Open \
    --tensor-parallel-size 4 \
    --max-model-len 262144 \
    --trust-remote-code

Using with SGLang

bash
python -m sglang.launch_server \
    --model-path prefeitura-rio/Rio-2.5-Open \
    --tp 4 \
    --context-length 262144 \
    --trust-remote-code

Model Details

Table

Developer	IplanRIO — Empresa Municipal de Informática e Planejamento S.A.
Base Model	Qwen3-30B-A3B-Thinking-2507
Architecture	Mixture-of-Experts (MoE) Transformer
Total Parameters	~30B
Active Parameters	~3B
Context Length	262,144 tokens
Default Max Output Length	81,920 tokens
Training Method	Distillation

Citation

If you use SwiReasoning, please also cite:

bibtex
@misc{shi2025swireasoning,
    title={SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs},
    author={Dachuan Shi et al.},
    year={2025},
    eprint={2510.05069},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Acknowledgments

Rio 2.5 Open 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.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Model Details

Model Provider

prefeitura-rio

Model Tree

Base

Qwen/Qwen3-30B-A3B-Thinking-2507

Fine-tuned

this model

Input Modalities

Text

Output Modalities

Text

Supported Functionality

Dedicated EndpointsContainer

Explore FriendliAI today

Get started Talk to an engineer

Key Features

30B total / 3B active parameters (Mixture-of-Experts)
262,144 token context window
SwiReasoning integration — dynamic explicit/latent reasoning switching for Pareto-superior accuracy and efficiency
Distilled from Qwen3-30B-A3B-Thinking-2507 with traces from Rio 2.5
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

Table with columns: Model, GPQA Diamond, LiveCodeBench, Composite Math*, AIME 2025, AIME 2026 I, HMMT 2025 I, HMMT 2025 II, BRUMO 2025, CMIMC 2025, SMT 2025
Model	GPQA Diamond	LiveCodeBench	Composite Math*	AIME 2025	AIME 2026 I	HMMT 2025 I	HMMT 2025 II	BRUMO 2025	CMIMC 2025	SMT 2025
Rio 2.5 Open	77.20%	69.60%	87.53%	93.33%	89.17%	83.33%	90.83%	88.33%	83.75%	83.96%
Rio 2.5 Open (w/o latent)	75.80%	69.40%	82.17%	90.00%	83.33%	76.67%	84.17%	85.83%	75.00%

*Composite Math is the average across all other mathematics benchmarks in this table.

Rio Model Family Comparison

Table with columns: Model, GPQA Diamond, LiveCodeBench, Composite Math*, AIME 2025
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%

Gains Over Base Model (Qwen3-30B-A3B-Thinking-2507)

Table with columns: Benchmark, Base Model, Rio 2.5 Open, Δ
Benchmark	Base Model	Rio 2.5 Open	Δ
GPQA Diamond	73.40%	77.20%	+3.80%
LiveCodeBench	66.00%	69.60%	+3.60%
Composite Math	76.08%	87.53%	+11.45%
AIME 2025	82.50%	93.33%

SwiReasoning: Latent/Explicit Reasoning

Rio 2.5 Open 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

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

python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "prefeitura-rio/Rio-2.5-Open"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)

prompt = "Write a poem about Rio de Janeiro."

messages = [
    {"role": "user", "content": prompt}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer([text], return_tensors="pt").to(model.device)

outputs = model.generate(
    **inputs,
    max_new_tokens=81920,
    temperature=0.6,
    top_p=0.95,
)

response = tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True)
print(response)

Using with vLLM

bash
vllm serve prefeitura-rio/Rio-2.5-Open \
    --tensor-parallel-size 4 \
    --max-model-len 262144 \
    --trust-remote-code

Using with SGLang

bash
python -m sglang.launch_server \
    --model-path prefeitura-rio/Rio-2.5-Open \
    --tp 4 \
    --context-length 262144 \
    --trust-remote-code

Model Details

Table

Developer	IplanRIO — Empresa Municipal de Informática e Planejamento S.A.
Base Model	Qwen3-30B-A3B-Thinking-2507
Architecture	Mixture-of-Experts (MoE) Transformer
Total Parameters	~30B
Active Parameters	~3B
Context Length	262,144 tokens
Default Max Output Length	81,920 tokens
Training Method	Distillation

Citation

If you use SwiReasoning, please also cite:

bibtex
@misc{shi2025swireasoning,
    title={SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs},
    author={Dachuan Shi et al.},
    year={2025},
    eprint={2510.05069},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Acknowledgments

Rio 2.5 Open 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.

Rio-2.5-Open

README

Key Features

Benchmark Results

Mathematics & STEM

Rio Model Family Comparison

Gains Over Base Model (Qwen3-30B-A3B-Thinking-2507)

SwiReasoning: Latent/Explicit Reasoning

How to Use

Using with vLLM

Using with SGLang

Model Details

Citation

Acknowledgments

Explore FriendliAI today

README

Key Features

Benchmark Results

Mathematics & STEM

Rio Model Family Comparison

Gains Over Base Model (Qwen3-30B-A3B-Thinking-2507)

SwiReasoning: Latent/Explicit Reasoning

How to Use

Using with vLLM

Using with SGLang

Model Details

Citation

Acknowledgments