Jan-v3-4B
A compact 4B model distilled from a larger teacher, with strong instruction following, built as a base you can fine-tune.
Overview
- Parameters
- 4B (3.6B non-embedding)
- Base model
- Qwen3-4B-Instruct-2507
- Context length
- 262,144 tokens (native)
- Architecture
- 36 layers, 32 Q heads / 8 KV heads (GQA)
- License
- Apache 2.0
18.0
Aider polyglot · % pass rate (strict)
Jan-v3-4B-base-instruct is a compact 4B parameter model created via post-training distillation from a larger teacher model. It is designed as an ownable, general-purpose base that is straightforward to fine-tune, with strong instruction following out of the box.
Try it in JanCapabilities
- Instruction following: Strong compliance with user instructions out of the box
- Fine-tuning base: Designed as a starting point for downstream fine-tuning
- Coding assistance: Effective lightweight coding help (a dedicated code variant, Jan-Code-4B, is also available)
- General purpose: Distillation preserves broad capabilities across standard benchmarks
Performance
Aider Benchmark (Polyglot Code Editing)
Jan-v3-4B leads its weight class on the Aider polyglot benchmark with a strict pass rate of 18, nearly 50% higher than the next best model (Qwen3-4B-Thinking at 12.1):
| Benchmark | Jan-v3-4B-base-instruct | Qwen3-4B-Thinking-2507 | Ministral-3-8B-Instruct-2512 | Olmo-3-7B-Think | Phi-4-Mini-Instruct | LFM2.5-1.2B-Thinking | LFM2-8B-A1B | Nemotron-Mini-4B-Instruct |
|---|---|---|---|---|---|---|---|---|
| Aider polyglot (Code editing · % pass rate (strict)) | 18.0 | 12.1 | 6.8 | 3.3 | 1.9 | 1.8 | 1.4 | 0.9 |
General Performance Comparison
Jan-v3-4B-base-instruct outperforms Qwen3-4B-Instruct-2507 (its base) across coding, math, and instruction-following benchmarks while remaining competitive on reasoning:
| Group | Benchmark | Qwen3-4B-Instruct-2507 | Jan-v3-4B-base-instruct |
|---|---|---|---|
| General reasoning | AIME 2025 | 0.47 | 0.47 |
| MMLU | 0.7260 | 0.7261 | |
| MMLU Pro | 0.6826 | 0.6734 | |
| IF-Bench (Prompt) | 0.3095 | 0.3265 | |
| IF-Bench (Instruct) | 0.3254 | 0.3463 | |
| EQBench | 0.8481 | 0.8469 | |
| Creative | 0.8049 | 0.7960 | |
| SimpleQA | 63.7 | 67.3 | |
| Math & STEM | GSM8K | 0.8870 | 0.9219 |
| GPQA | 0.4554 | 0.4551 | |
| GPQA-Diamond | 0.50 | 0.5101 | |
| Coding | Aider-polyglot | 12.9 | 18.0 |
Requirements
- Memory:
- Minimum: 8GB RAM (with Q4 quantization)
- Recommended: 16GB RAM (with Q8 quantization)
- Hardware: CPU or GPU
- API Support: OpenAI-compatible at localhost:1337
Using Jan-v3-4B
Quick Start
- Download Jan Desktop
- Click Open in Jan above, or open the Hub and search for Jan-v3-4B-base-instruct
- Choose a variant that fits your hardware, then click Download
- Once downloaded, select Jan-v3-4B-base-instruct from the model selector in any chat
- Start chatting — no additional configuration needed
Demo
Try it live at chat.jan.ai (opens in a new tab)
Deployment Options
Using vLLM:
vllm serve janhq/Jan-v3-4B-base-instruct \ --host 0.0.0.0 \ --port 1234 \ --enable-auto-tool-choice \ --tool-call-parser hermes
Using llama.cpp:
llama-server --model Jan-v3-4B-base-instruct-Q8_0.gguf \ --host 0.0.0.0 \ --port 1234 \ --jinja \ --no-context-shift
Recommended Parameters
temperature: 0.7top_p: 0.8top_k: 20
What Jan-v3-4B Does Well
- Long context: Native 262k token context window for handling large documents
- Instruction following: Reliable and consistent response to user instructions
- Fine-tuning: Designed to be a strong, ownable starting point for specialization
- General tasks: Broad capabilities preserved through distillation from a larger model
Limitations
- Model size: 4B parameters limit complex multi-step reasoning compared to larger models
- Specialized domains: For coding specifically, consider Jan-Code-4B
Downloads
- Jan: Download Jan
- Jan-v3-4B-base-instruct: Hugging Face (opens in a new tab) GGUF (opens in a new tab)
GGUF quantizations
- Q4_K_M: Good balance of size and quality
- Q5_K_M: Better quality, slightly larger
- Q8_0: Highest quality quantization (recommended)
Talk to us
- Discussions: Hugging Face community (opens in a new tab)
- Chat: Discord (opens in a new tab)
- Issues: GitHub (opens in a new tab)



