Instructions to use PrimeIntellect/INTELLECT-3-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PrimeIntellect/INTELLECT-3-FP8 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="PrimeIntellect/INTELLECT-3-FP8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PrimeIntellect/INTELLECT-3-FP8")
model = AutoModelForCausalLM.from_pretrained("PrimeIntellect/INTELLECT-3-FP8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use PrimeIntellect/INTELLECT-3-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PrimeIntellect/INTELLECT-3-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PrimeIntellect/INTELLECT-3-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PrimeIntellect/INTELLECT-3-FP8

SGLang

How to use PrimeIntellect/INTELLECT-3-FP8 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "PrimeIntellect/INTELLECT-3-FP8" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PrimeIntellect/INTELLECT-3-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "PrimeIntellect/INTELLECT-3-FP8" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PrimeIntellect/INTELLECT-3-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PrimeIntellect/INTELLECT-3-FP8 with Docker Model Runner:
```
docker model run hf.co/PrimeIntellect/INTELLECT-3-FP8
```

Jackmin108 commited on Nov 27, 2025

Commit

1cb0a84

verified ·

1 Parent(s): b10802f

Update README.md

Browse files

Files changed (1) hide show

README.md +34 -9

README.md CHANGED Viewed

@@ -4,6 +4,10 @@ tags:
 - prime-rl
 - verifiers
 - prime-intellect
 license: mit
 language:
 - en
@@ -14,24 +18,45 @@ pipeline_tag: text-generation
 # INTELLECT-3
 **INTELLECT-3** is a 106B (A12B) parameter Mixture-of-Experts reasoning model post-trained from [GLM-4.5-Air-Base](https://huggingface.co/zai-org/GLM-4.5-Air-Base) using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).
-Training was performed with [prime-rl](https://github.com/PrimeIntellect-ai/prime-rl) using environments built with the [verifiers](https://github.com/PrimeIntellect-ai/verifiers) library. All training and evaluation environments are available on the [Environments Hub](https://app.primeintellect.ai/dashboard/environments).
 The model, training frameworks, and environments are open-sourced under fully-permissive licenses (MIT and Apache 2.0).
-For more details, see the [technical report](PAPER_LINK_PLACEHOLDER).
 ## Evaluation
 INTELLECT-3 achieves best-in-class performance on math, coding, and reasoning benchmarks:
-| Benchmark | Score |
-|-----------|-------|
-| AIME 2025 | 88.0 |
-| LiveCodeBench v6 | 69.3 |
-| GPQA Diamond | 74.4 |
-| HLE | 14.6 |
 ## Model Variants
@@ -67,4 +92,4 @@ vllm serve PrimeIntellect/INTELLECT-3-FP8 \
   year={2025},
   url={https://huggingface.co/PrimeIntellect/INTELLECT-3}
 }
-```

 - prime-rl
 - verifiers
 - prime-intellect
+- reinforcement-learning
+- reasoning
+- agentic
+- mixture-of-experts
 license: mit
 language:
 - en
 # INTELLECT-3
+<div align="center">
+<img src="pi_logo.png" alt="Prime Intellect Logo" width="100px"/>
+</div>
+<p align="center">
+    🚀 <strong>State-of-the-art 100B+ parameter Mixture-of-Experts model trained with large-scale reinforcement learning</strong>
+    <br><br>
+    Trained with <a href="https://github.com/PrimeIntellect-ai/prime-rl">prime-rl</a> infra and <a href="https://github.com/PrimeIntellect-ai/verifiers">verifiers</a> environments
+    <br>
+    Environments on <a href="https://app.primeintellect.ai/dashboard/environments">Environments Hub</a> | Read the <a href="https://primeintellect.ai/blog/intellect-3">Blog</a> & <a href="https://huggingface.co/PrimeIntellect/INTELLECT-3">Technical Report</a>
+    <br>
+    <a href="https://discord.gg/RC5GvMbfDf">X</a>  | <a href="https://discord.gg/RC5GvMbfDf">Discord</a> | <a href="https://app.primeintellect.ai/dashboard/create-cluster">Prime Compute Platform</a>
+</p>
+## Introduction
 **INTELLECT-3** is a 106B (A12B) parameter Mixture-of-Experts reasoning model post-trained from [GLM-4.5-Air-Base](https://huggingface.co/zai-org/GLM-4.5-Air-Base) using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).
+![bench](temp_bench.png)
+Training was performed with [prime-rl](https://github.com/PrimeIntellect-ai/prime-rl) using environments built with the [verifiers](https://github.com/PrimeIntellect-ai/verifiers) library.
+All training and evaluation environments are available on the [Environments Hub](https://app.primeintellect.ai/dashboard/environments).
 The model, training frameworks, and environments are open-sourced under fully-permissive licenses (MIT and Apache 2.0).
+For more details, see the [technical report](https://huggingface.co/PrimeIntellect/INTELLECT-3).
 ## Evaluation
 INTELLECT-3 achieves best-in-class performance on math, coding, and reasoning benchmarks:
+| Benchmark | MATH-500 | AIME24 | AIME25 | LCB | GPQA | HLE | MMLU-Pro |
+|-----------|----------|---------|---------|--------|------|-----|----------|
+| INTELLECT-3 | **98.1** | **90.8** | **88.0** | 69.3 | 74.4 | 14.6 | 81.9 |
+| GLM-4.5-Air | 97.8 | 84.6 | 82.0 | 61.5 | 73.3 | 13.3 | 73.9 |
+| GLM-4.5 | 97.0 | 85.8 | 83.3 | 64.5 | 77.0 | X | X |
+| DeepSeek R1 0528 | 87.3 | 83.2 | 73.4 | 62.5 | 77.5 | 15.9 | 75.3 |
+| DeepSeek v3.2 | 96.8 | 88.1 | 84.7 | **71.6** | **81.4** | **17.9** | **84.6** |
+| GPT-O5S 120B | 96.0 | 75.8 | 77.7 | 69.9 | 70.0 | 10.6 | 67.1 |
 ## Model Variants
   year={2025},
   url={https://huggingface.co/PrimeIntellect/INTELLECT-3}
 }
+```