Rnj-1

Changelog

• Update December 20, 2025:
  • System prompt and temperature recommendations: Resolve premature truncations and mitigate unprompted code outputs.
  • Updates to default chat template.
  • Updated evaluation results.
  • Links to model generations for evals.
  • Instructions for long-context extrapolation.
• Initial version: December 8, 2025

Capabilities

Benchmark Results

Base Model rnj-1

Instruct Model rnj-1-instruct

Rnj-1 models are designed to be extended

Highlights of abilities

• Code generation: Both rnj-1-instruct and rnj-1 demonstrate strong code generation abilities as measured on tasks like HumanEval+, MBPP+, BigCodeBench, and LiveCodeBench v6. Both models compete with the strongest open weight models, sometimes outperforming even larger models such as GPT OSS 20B. We measured code comprehension abilities using the task of predicting inputs given outputs and vice-versa, Crux-IO. We find our models outperform comparable baselines. For multi-lingual code generation capabilities across programming languages we measure MultiPL-E on 6 languages (C++, TypeScript, Java, JavaScript, Shell, PHP) and we find performance close to the strongest model.
• Agentic and Tool Use: rnj-1-instruct dominates the pack on agentic coding, one of our target abilities. SWE-bench performance is indicative of the model’s ability to tackle everyday software engineering tasks. The model is an order of magnitude stronger than comparably sized models on SWE-bench and approaches the capabilities available in much larger models. It scores 20.8% on SWE-bench Verified in bash-only mode, which is higher than Gemini 2.0 flash and Qwen2.5-Coder 32B Instruct under the same agentic framework (leaderboard).
  
  There is a surge of interest in developing models’ abilities to write performant code. rnj-1-instruct is able to use a profiler to iteratively improve the performance of the code it writes. For instance, on Enamel, which measures abilities to write efficient solutions to algorithmic problems, the model outperforms all other models under the same setting.
  
  Furthermore, rnj-1-instruct surpasses comparable models in tool use performance as measured by the Berkeley Functional Calling Leaderboard (BFCL).
• Code Infilling : Having specifically been trained on FIM-ed pre-training data, rnj-1 exhibits strong infilling abilities, which have been further enhanced during post-training. The base model rnj-1 scores highly on HE-FIM-Python (avg) at 82.49% and rnj-1-instruct achieves 86.21%.
• Mathematical Problem Solving: rnj-1-instruct shows strong mathematical abilities across several levels of difficulty from elementary math (GSM8k), high school and undergraduate math (Minerva-MATH), and competition math (AIME ‘24 and ‘25). On harder subjects, it outcompetes or is on par with the strongest model in the pack.
• Scientific Reasoning: rnj-1-instruct exhibits long-context reasoning abilities that are needed to solve hard science and technical questions in GPQA-Diamond and SuperGPQA.

Demos: Rnj-1 models generalize to unseen tasks

• Coding assistant: rnj-1-instruct can operate in agentic mode to create a playable game in a single shot inside of Cline: screen recording.
• Agentic use: rnj-1-instruct functions seamlessy within the agentic framework of mini-swe-agent. Given a task such as fixing an issue described in a pull request (PR), fixing a security vulnerability, or writing performant code, it is able to reason across its full context across multiple turns to solve the task. These lead to “trajectories” which are pairs of “Assistant” and “User” turns. Here are a few recordings that show the model’s reasoning abilities across these turns: 1) a SWE task of identifying coding convention violation: screen recording, 2) fixing a security vulnerability: screen recording, 3) diagnosing code performance bottlenecks by running a profiler in the environment and iteratively improving the code: screen recording.
• Data analysis in an interactive chat: rnj-1-instruct can work in interactive chat mode to solve a data analysis and visualization task: screen recording.

Architecture

Training Dynamics

• Warmup: Linear ramp-up from 0 to 2e-3 over the first 5K steps.
• Stable phase: Constant learning rate of 2e-3 from 5K → 230K steps.
• Decay: Cosine decay from 2e-3 → 2e-5 from 230K → 380K steps.
• Final stable phase: Constant 2e-5 learning rate from 380K → 443.5K steps, concluding pre-training.

• 18M tokens for pre-training.
• 24M tokens for mid-training.
• 16M tokens for SFT.

Long-Context Extrapolation (up to 128k)

Recommendations

System Prompt & Temperature

Propensity to write code

How to use

Serverless API and online playgrounds

• Together.AI: Rnj-1 Instruct is available via API on the Together.ai model platform for serverless inference. It’s also available in the Together.ai playground for quick and easy experimentation.
• HuggingFace: Rnj-1 Instruct is also hosted via Hugging Face Spaces.

Running Rnj-1 locally

Running Rnj-1 on your laptop with llama.cpp

Use with transformers

1. Example code for querying model without tools
   python

   1import torch 2from transformers import AutoTokenizer, AutoModelForCausalLM 3import os 4 5model_id = "EssentialAI/rnj-1-instruct" 6os.environ["HF_TOKEN"] = <YOUR-HF-TOKEN> 7 8print(f"Loading model: {model_id}...") 9model = AutoModelForCausalLM.from_pretrained( 10 model_id, 11 dtype=torch.bfloat16, 12 device_map="auto", 13) 14tokenizer = AutoTokenizer.from_pretrained(model_id) 15 16print("Model and tokenizer loaded successfully.") 17 18messages = [ 19 {"role": "system", "content": "You are a helpful AI assistant."}, # Optional system message 20 {"role": "user", "content": "Who are you?"} 21] 22 23input_ids = tokenizer.apply_chat_template( 24 messages, 25 add_generation_prompt=True, 26 return_tensors="pt" 27).to(model.device) 28 29# --- Generate Prediction --- # 30print("Generating prediction...") 31output_ids = model.generate( 32 input_ids, 33 max_new_tokens=50, 34 pad_token_id=tokenizer.eos_token_id, 35 do_sample=True, 36 temperature=0.2, 37 top_p=0.95 38) 39 40response = tokenizer.decode(output_ids[0][input_ids.shape[-1]:], skip_special_tokens=True) 41print(response)
2. Example code for querying with tools
   Rnj-1 supports tool-calling which can be parsed by hermes tool-call parser. The tool calls are formatted inside <tool_call> and </tool_call> tags. An example usage is as follows:
   python

   1tools = [ 2 { 3 "type": "function", 4 "function": { 5 "name": "get_weather", 6 "description": "Get the current weather in a given location", 7 "parameters": { 8 "type": "object", 9 "properties": { 10 "location": {"type": "string", "description": "City and state, e.g., 'San Francisco, CA'"}, 11 "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]} 12 }, 13 "required": ["location", "unit"], 14 }, 15 }, 16 }, 17] 18 19messages = [ 20 {"role": "system", "content": "You are a helpful AI assistant."}, # Optional system message 21 {"role": "user", "content": "What is the weather in San Francisco, CA in Celsius?"} 22] 23 24input_ids = tokenizer.apply_chat_template( 25 messages, 26 tools=tools, 27 add_generation_prompt=True, 28 return_tensors="pt" 29).to(model.device) 30 31# --- Generate Prediction --- # 32print("Generating prediction...") 33output_ids = model.generate( 34 input_ids, 35 max_new_tokens=200, 36 pad_token_id=tokenizer.eos_token_id, 37 do_sample=True, 38 temperature=0.2, 39 top_p=0.95 40) 41 42response = tokenizer.decode(output_ids[0][input_ids.shape[-1]:], skip_special_tokens=False) 43# NOTE: skip_special_tokens is set to False. 44print(response)
3. Example code for fill-in-the-middle (FIM)
   Rnj-1 supports FIM, we show an example payload to trigger FIM mode for Rnj-1 below:
   python

   1PRE = "<|pre_fim|>" 2MID = "<|mid_fim|>" 3SUF = "<|suf_fim|>" 4 5prefix = """def binary_search(arr, target): 6 lo = 0 7 hi = len(arr) - 1 8 9 while lo <= hi: 10""" 11 12suffix = """ 13 return -1 14""" 15 16input = PRE + prefix + SUF + suffix + MID 17 18messages = [ 19 {"role": "system", "content": "You are a helpful AI assistant."}, 20 {"role": "user", "content": input} 21] 22 23input_ids = tokenizer.apply_chat_template( 24 messages, 25 tools=tools, 26 add_generation_prompt=True, 27 return_tensors="pt" 28).to(model.device) 29 30# --- Generate Prediction --- # 31print("Generating prediction...") 32output_ids = model.generate( 33 input_ids, 34 max_new_tokens=100, 35 pad_token_id=tokenizer.eos_token_id, 36 do_sample=True, 37 temperature=0.2, 38 top_p=0.95 39) 40 41response = tokenizer.decode(output_ids[0][input_ids.shape[-1]:], skip_special_tokens=False) 42print(response)

Serving Rnj-1 on GPUs

vLLM

SGLang

IDEs and Agents: Claude Code, Cline, Mini-SWE-Agent

Use with Cline

Use with Claude Code

Agentic mode with Mini-SWE-Agent

Known limitations

Hallucinations and factual inaccuracies

Identity and knowledge cutoff

License

Contact

Citation