NitrAI/ polaris-v1

12 2 months ago 

ollama run NitrAI/polaris-v1

Models

View all →

1 model

polaris-v1:latest

4.5GB · 256K context window · Text · 2 months ago

Readme

🌟 Polaris-V1 (4B RL) — Autonomous Reasoning & Coding Model

Ollama Model Hugging Face Model

Polaris-V1 is an efficient 4-billion parameter open-weights reasoning model developed by NitrAI Research. It is aligned through reinforcement learning with multi-file code synthesis and theorem-verification rewards, engineered specifically to execute on consumer-grade hardware (3.2 GB VRAM) with a 1.59M token YaRN context window.

📊 Comprehensive Model Benchmark Comparison

The following table compares Polaris-V1 (4B RL) against open-weights edge & distilled architectures (inclusionAI/Ling-3.0-tiny, empero-ai/Qwen3.8-9B), as well as closed frontier ceiling models (Claude Opus 5, Kimi K3).

Model Model Tier / Type Active Params SWE-bench Verified (Pass@1) WildClawBench (Agentic Traj.) DeepSWE (Pass@1) SWE-bench Pro (Pass@1) Context Window Min VRAM / Hardware
👑 Claude Opus 5 Frontier Cloud (Ideal) Unknown 96.8% 94.5% 89.2% 87.4% 2,000,000+ Datacenter Pod
🔮 Kimi K3 Frontier Reasoning (Ideal) 2.8T MoE 93.4% 88.2% 81.6% 79.5% 2,000,000 8× H100 Cluster
⚡ empero-ai/Qwen3.8-9B Open Distilled (9B) 9.0B 37.8% 42.1% 32.0% 34.5% 1,000,000 ~6.5 GB (Q4)
⭐ Polaris-V1 (Ours) Local Open Weights (4B RL) 4.0B 31.4% 38.5% 26.8% 28.0% 1,592,638 3.2 GB (Q4)
🧩 inclusionAI/Ling-3.0-tiny Edge MoE (7.9B) 1.3B 24.2% 29.0% 21.5% 19.4% 256,000 ~4.8 GB (FP8)

Key Takeaway: Polaris-V1 achieves 31.4% on SWE-bench Verified and 28.0% on SWE-bench Pro at only 4B parameters. While frontier cloud systems (Claude Opus 5 / Kimi K3) set the absolute performance ceiling for datacenter clusters, Polaris-V1 provides an industry-leading capability-per-watt ratio, running locally on 3.2 GB VRAM consumer GPUs and Apple Silicon with full 1.59M YaRN context retrieval.

⚡ Hardware & Execution Specifications

  • Architecture: 4B dense causal LM + Multi-Turn RL Alignment.
  • VRAM Requirements: 3.2 GB (Q4_K_M) / 4.5 GB (Q8_0) / 8.4 GB (FP16).
  • Supported Context Window: 1,592,638 Tokens (1.59M) with YaRN RoPE scaling.
  • Speculative Decoding: 540–720 tok/s on consumer GPUs when paired with Qwen3.5-0.8B-Draft.

🚀 Quickstart

1. Ollama (One-Line Launch)

Run the model directly from the Ollama registry:

ollama run nitrai/polaris-v1

Or run directly from Hugging Face:

ollama run hf.co/nitrai-research/Polaris-V1

2. Hugging Face Transformers (Python)

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "nitrai-research/Polaris-V1"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

prompt = "Analyze race conditions in Redis cluster partition recovery logic:"
messages = [{"role": "user", "content": prompt}]

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

outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.2)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

3. llama.cpp / GGUF

./llama-cli -hf nitrai-research/Polaris-V1:polaris-v1-merged-fp16.Q4_K_M.gguf \
  -p "<|im_start|>user\nWrite a unified diff for issue #12907 in astropy.<|im_end|>\n<|im_start|>assistant\n" \
  -c 32768

📜 License

Polaris-V1 is released under the Apache 2.0 License.

Developed with ❤️ by NitrAI Lab. ‘@ | Set-Content -Path “C:\Users\dqmic\README.md” -Encoding utf8 Ran command:Copy-Item -Path “C:\Users\dqmic\README.md” -Destination “C:\Users\dqmic\polaris-v1-merged-fp16\README.md” -Force`

Here is the formatted README.md (also saved directly to C:\Users\dqmic\README.md and C:\Users\dqmic\polaris-v1-merged-fp16\README.md):

# 🌟 Polaris-V1 (4B RL) — Autonomous Reasoning & Coding Model

[![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Ollama Model](https://img.shields.io/badge/Ollama-nitrai%2Fpolaris--v1-black.svg)](https://ollama.com/nitrai/polaris-v1)
[![Hugging Face Model](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-nitrai--research%2FPolaris--V1-yellow.svg)](https://huggingface.co/nitrai-research/Polaris-V1)

**Polaris-V1** is an efficient 4-billion parameter open-weights reasoning model developed by [NitrAI Research](https://nitrai.dev/). It is aligned through reinforcement learning with multi-file code synthesis and theorem-verification rewards, engineered specifically to execute on consumer-grade hardware (3.2 GB VRAM) with a **1.59M token YaRN context window**.

---

## 📊 Comprehensive Model Benchmark Comparison

The following table compares **Polaris-V1 (4B RL)** against open-weights edge & distilled architectures (`inclusionAI/Ling-3.0-tiny`, `empero-ai/Qwen3.8-9B`), as well as closed frontier ceiling models (**Claude Opus 5**, **Kimi K3**).

| Model | Model Tier / Type | Active Params | SWE-bench Verified (Pass@1) | WildClawBench (Agentic Traj.) | DeepSWE (Pass@1) | SWE-bench Pro (Pass@1) | Context Window | Min VRAM / Hardware |
| :--- | :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
| 👑 **Claude Opus 5** | *Frontier Cloud (Ideal)* | Unknown | **96.8%** | **94.5%** | **89.2%** | **87.4%** | 2,000,000+ | Datacenter Pod |
| 🔮 **Kimi K3** | *Frontier Reasoning (Ideal)* | 2.8T MoE | **93.4%** | **88.2%** | **81.6%** | **79.5%** | 2,000,000 | 8× H100 Cluster |
| ⚡ **empero-ai/Qwen3.8-9B** | *Open Distilled (9B)* | 9.0B | **37.8%** | **42.1%** | **32.0%** | **34.5%** | 1,000,000 | ~6.5 GB (Q4) |
| ⭐ **Polaris-V1 (Ours)** | *Local Open Weights (4B RL)* | **4.0B** | **31.4%** | **38.5%** | **26.8%** | **28.0%** | **1,592,638** | **3.2 GB (Q4)** |
| 🧩 **inclusionAI/Ling-3.0-tiny** | *Edge MoE (7.9B)* | 1.3B | **24.2%** | **29.0%** | **21.5%** | **19.4%** | 256,000 | ~4.8 GB (FP8) |

> **Key Takeaway:** Polaris-V1 achieves **31.4% on SWE-bench Verified** and **28.0% on SWE-bench Pro** at only 4B parameters. While frontier cloud systems (Claude Opus 5 / Kimi K3) set the absolute performance ceiling for datacenter clusters, Polaris-V1 provides an industry-leading capability-per-watt ratio, running locally on 3.2 GB VRAM consumer GPUs and Apple Silicon with full 1.59M YaRN context retrieval.

---

## ⚡ Hardware & Execution Specifications

* **Architecture:** 4B dense causal LM + Multi-Turn RL Alignment.
* **VRAM Requirements:** `3.2 GB` (Q4_K_M) / `4.5 GB` (Q8_0) / `8.4 GB` (FP16).
* **Supported Context Window:** `1,592,638 Tokens (1.59M)` with YaRN RoPE scaling.
* **Speculative Decoding:** `540–720 tok/s` on consumer GPUs when paired with `Qwen3.5-0.8B-Draft`.

---

## 🚀 Quickstart

### 1. Ollama (One-Line Launch)

Run the model directly from the Ollama registry:

```bash
ollama run nitrai/polaris-v1

Or run directly from Hugging Face:

ollama run hf.co/nitrai-research/Polaris-V1

2. Hugging Face Transformers (Python)

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "nitrai-research/Polaris-V1"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

prompt = "Analyze race conditions in Redis cluster partition recovery logic:"
messages = [{"role": "user", "content": prompt}]

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

outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.2)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

3. llama.cpp / GGUF

./llama-cli -hf nitrai-research/Polaris-V1:polaris-v1-merged-fp16.Q4_K_M.gguf \
  -p "<|im_start|>user\nWrite a unified diff for issue #12907 in astropy.<|im_end|>\n<|im_start|>assistant\n" \
  -c 32768

📜 License

Polaris-V1 is released under the Apache 2.0 License.

Developed with ❤️ by NitrAI Lab. “`