Instructions to use xupy21/ContextRL_Klear_AgentForge_8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xupy21/ContextRL_Klear_AgentForge_8B with Transformers:

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

pipe = pipeline("text-generation", model="xupy21/ContextRL_Klear_AgentForge_8B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("xupy21/ContextRL_Klear_AgentForge_8B")
model = AutoModelForMultimodalLM.from_pretrained("xupy21/ContextRL_Klear_AgentForge_8B")
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 Settings

vLLM

How to use xupy21/ContextRL_Klear_AgentForge_8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xupy21/ContextRL_Klear_AgentForge_8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xupy21/ContextRL_Klear_AgentForge_8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/xupy21/ContextRL_Klear_AgentForge_8B

SGLang

How to use xupy21/ContextRL_Klear_AgentForge_8B 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 "xupy21/ContextRL_Klear_AgentForge_8B" \
    --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": "xupy21/ContextRL_Klear_AgentForge_8B",
		"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 "xupy21/ContextRL_Klear_AgentForge_8B" \
        --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": "xupy21/ContextRL_Klear_AgentForge_8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use xupy21/ContextRL_Klear_AgentForge_8B with Docker Model Runner:
```
docker model run hf.co/xupy21/ContextRL_Klear_AgentForge_8B
```

ContextRL_Klear_AgentForge_8B

File size: 2,227 Bytes

---
license: apache-2.0
base_model:
  - Klear-AgentForge-8B
pipeline_tag: text-generation
library_name: transformers
tags:
  - agentic
  - code
  - software-engineering
  - reinforcement-learning
  - grpo
  - context-aware
  - long-context

---

# ContextRL-Klear-AgentForge-8B

This is the **agentic (long-horizon) model** released with the paper
**Context-Aware RL for Agentic and Multimodal LLMs**.
It is fine-tuned from **Klear-AgentForge-8B**, a model specialized for complex agentic
coding, using **ContextRL**, a context-aware reinforcement learning method that augments
standard GRPO with an auxiliary *context-selection* objective to improve fine-grained
context grounding in long-horizon agent trajectories.

## Results

Across 5 long-horizon benchmarks (2 in-distribution agentic coding, 3 out-of-distribution),
ContextRL improves over the standard GRPO baseline by **+3.2 points** on average, while
improving every individual benchmark.

| Benchmark              | Base | RL (GRPO) | ContextRL (Ours) |
| ---------------------- | ---- | --------- | ---------------- |
| SWE-Bench Verified     | 26.6 | 28.0      | **30.2**         |
| SWE-Bench Lite         | 21.0 | 21.7      | **24.0**         |
| LiveCodeBench v6       | 21.7 | 22.3      | **24.0**         |
| LongBench v2 (Overall) | 27.4 | 27.0      | **29.6**         |
| LongBench v2 (Long)    | 21.3 | 24.1      | **28.7**         |
| NIAH                   | 68.3 | 65.5      | **71.3**         |

*Metrics: SWE-Bench Verified/Lite resolve rate (%), LiveCodeBench v6 solve rate (%), LongBench v2 accuracy (%), NIAH mean recall (%).* On the long-context tasks (LongBench v2, NIAH) where standard outcome-based GRPO struggles or regresses, ContextRL surpasses both the base model and the RL baseline, demonstrating strong out-of-distribution generalization.

## Usage

This model follows the same interface as its Klear-AgentForge-8B base and can be loaded
with `transformers`. Training and evaluation code, data construction pipelines, and
detailed configurations are available in the repository:
👉 **https://github.com/xupy2003/ContextAwareRL**
Please refer to the repo's README for environment setup, inference scripts, and
reproduction instructions.