Bifrost-AI/Mirage-OpenReasoning-Nemotron-7B-GGUF

Mirage OpenReasoning Nemotron 7B

This fine-tuned variant of the OpenReasoning Nemotron 7B model was supervised fine-tuned on blockchain-specific datasets(Bifrost-AI/Solana-Vanguard-Challenge), optimized for downstream tasks in blockchain coding and smart contract development on the Solana ecosystem.

The Solana Vanguard Challenge dataset, comprising 1,000 diverse and in-depth questions, offers full-spectrum coverage of the Solana ecosystem. It spans fundamental blockchain concepts, advanced on-chain programming in Rust and the Anchor framework, client-side integration in TypeScript, detailed security strategies, and performance as well as regulatory considerations.

Mirage OpenReasoning Nemotron 7B is in active development with additional fine-tuning sessions, & benchmark statistics coming soon!

Provided Quants

Link	Type	Size/GB	Notes
GGUF	Q4_K_S	4	fast, recommended
GGUF	Q5_K-S	5
GGUF	Q6_K	6	very good quality
GGUF	Q8_0	8	fast, best quality
GGUF	F16	15	16 bpw, highest quality

Training Session:

• Time: 10 hours & 48 minutes
• GPU: NVIDIA GeForce RTX 3090
• Batches: 500
• Context-Size: 2043
• Batch-size: 1
• Learning-rate: 2e-5
• Training-loss: 1.25
• Eval-loss: 0.95

Dataset Composition

• Total Questions: 1,000
• Languages Covered:
  • Rust: On-chain smart contract development, security best practices, advanced state management, CPIs, PDAs, and more.
  • TypeScript: Client-side integration using @solana/web3.js, wallet adapters, Metaplex for NFT protocols, dynamic transaction composition, and front-end dApp development.
• Planned Extensions:
  • C# (Solnet): To be integrated later for .NET ecosystem coverage.

How to use the models?

To run inference on coding problems:

import transformers
import torch
model_id = "Bifrost-AI/Mirage-OpenReasoning-Nemotron-7B"
pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)
# Code generation prompt
prompt = """You are a helpful and harmless assistant. You should think step-by-step before responding to the instruction below.
Please use python programming language only.
You must use ```python for just the final solution code block with the following format:
```python
# Your code here
```
{user}
"""

# Math generation prompt
# prompt = """Solve the following math problem. Make sure to put the answer (and only answer) inside \\boxed{}.
# 
# {user}
# """

# Science generation prompt
# You can refer to prompts here -
# https://github.com/NVIDIA/NeMo-Skills/blob/main/nemo_skills/prompt/config/generic/hle.yaml (HLE)
# https://github.com/NVIDIA/NeMo-Skills/blob/main/nemo_skills/prompt/config/eval/aai/mcq-4choices-boxed.yaml (for GPQA)
# https://github.com/NVIDIA/NeMo-Skills/blob/main/nemo_skills/prompt/config/eval/aai/mcq-10choices-boxed.yaml (MMLU-Pro)

messages = [
    {
        "role": "user",
        "content": prompt.format(user="Write a program to calculate the sum of the first $N$ fibonacci numbers")},
]
outputs = pipeline(
    messages,
    max_new_tokens=64000,
)
print(outputs[0]["generated_text"][-1]['content'])

--------------------------Base Model Card--------------------------

OpenReasoning-Nemotron-7B Overview

Description:

OpenReasoning-Nemotron-7B is a large language model (LLM) which is a derivative of Qwen2.5-7B-Instruct (AKA the reference model). It is a reasoning model that is post-trained for reasoning about math, code and science solution generation. We evaluated this model with up to 64K output tokens. The OpenReasoning model is available in the following sizes: 1.5B, 7B and 14B and 32B.

This model is ready for commercial/non-commercial research use.

License/Terms of Use:

GOVERNING TERMS: Use of the models listed above are governed by the Creative Commons Attribution 4.0 International License (CC-BY-4.0). ADDITIONAL INFORMATION: Apache 2.0 License

Scores on Reasoning Benchmarks

Our models demonstrate exceptional performance across a suite of challenging reasoning benchmarks. The 7B, 14B, and 32B models consistently set new state-of-the-art records for their size classes.

Model	AritificalAnalysisIndex*	GPQA	MMLU-PRO	HLE	LiveCodeBench*	SciCode	AIME24	AIME25	HMMT FEB 25
1.5B	31.0	31.6	47.5	5.5	28.6	2.2	55.5	45.6	31.5
7B	54.7	61.1	71.9	8.3	63.3	16.2	84.7	78.2	63.5
14B	60.9	71.6	77.5	10.1	67.8	23.5	87.8	82.0	71.2
32B	64.3	73.1	80.0	11.9	70.2	28.5	89.2	84.0	73.8

* This is our estimation of the Artificial Analysis Intelligence Index, not an official score.

* LiveCodeBench version 6, date range 2408-2505.

Combining the work of multiple agents

OpenReasoning-Nemotron models can be used in a "heavy" mode by starting multiple parallel generations and combining them together via generative solution selection (GenSelect). To add this "skill" we follow the original GenSelect training pipeline except we do not train on the selection summary but use the full reasoning trace of DeepSeek R1 0528 671B instead. We only train models to select the best solution for math problems but surprisingly find that this capability directly generalizes to code and science questions! With this "heavy" GenSelect inference mode, OpenReasoning-Nemotron-32B model surpasses O3 (High) on math and coding benchmarks.

Model	Pass@1 (Avg@64)	Majority@64	GenSelect
1.5B
AIME24	55.5	76.7	76.7
AIME25	45.6	70.0	70.0
HMMT Feb 25	31.5	46.7	53.3
7B
AIME24	84.7	93.3	93.3
AIME25	78.2	86.7	93.3
HMMT Feb 25	63.5	83.3	90.0
LCB v6 2408-2505	63.4	n/a	67.7
14B
AIME24	87.8	93.3	93.3
AIME25	82.0	90.0	90.0
HMMT Feb 25	71.2	86.7	93.3
LCB v6 2408-2505	67.9	n/a	69.1
32B
AIME24	89.2	93.3	93.3
AIME25	84.0	90.0	93.3
HMMT Feb 25	73.8	86.7	96.7
LCB v6 2408-2505	70.2	n/a	75.3
HLE	11.8	13.4	15.5

---