Qwen 2.5 Coder 32B Instruct
Qwen · Apache 2.0
The coding model that defines the builder workflow. Matches GPT-4 on HumanEval. This is what Cursor and Continue.dev users run locally when they want to eliminate API dependency. Apache 2.0 license. The cornerstone of the 'Full AI Builder' profile.
- Parameters
- 32.5B
- Architecture
- Dense
- Context
- 131,072 tokens
- Released
- 2024-11-12
- Engines
- llama.cpp, ollama, vLLM
- Builder Tools
- Cursor, Continue, Aider, Windsurf, Codex CLI
Parameters
32.5B
VRAM
21.9 GB
Context
128K
Formats
5
GPUs
39
Qwen 2.5 Coder 32B Instruct (32.5B) requires 21.9 GB VRAM at recommended quality (Q5_K_M). At efficient quality (Q4_K_M), it fits in 18.4 GB VRAM, making it compatible with the NVIDIA RTX 4090 Laptop (150-175W). On NVIDIA Grace Blackwell Ultra GB300, expect approximately 140 tok/s at Q5_K_M. For the best experience, AMD AI Powerhouse ($1,818) is recommended.
Source: OwnRig methodology
21.9 GB
Q5_K_M
19.5 GB
128K tokens
Coding
VRAM Requirements
| Quality | Quantization | VRAM | File Size |
|---|---|---|---|
| full | Q8_0 | 34.6 GB | 32.5 GB |
| recommended | Q5_K_M | 21.9 GB | 19.5 GB |
| efficient | Q4_K_M | 18.4 GB | 16.3 GB |
| compressed | Q3_K_M | 14.8 GB | 12.7 GB |
| compressed | Q2_K | 11.6 GB | 9.8 GB |
Context Length Impact
KV cache VRAM at Q5_K_M quality. Longer context = more memory.
| Context | KV Cache | Total VRAM |
|---|---|---|
| 2K | 410 MB | 22.3 GB |
| 4K | 819 MB | 22.7 GB |
| 8K | 1.5 GB | 23.4 GB |
| 16K | 3.1 GB | 25 GBexceeds 24 GB |
| 32K | 6.1 GB | 28 GBexceeds 24 GB |
| 64K | 12.3 GB | 34.2 GBexceeds 24 GB |
| 128K | 24.6 GB | 46.5 GBexceeds 24 GB |
Compatible GPUs
39 devicesShowing 39 of 39 entries
Builder Context
Qwen 2.5 Coder 32B Instruct is commonly used with Cursor, Continue, Aider, Windsurf, Codex CLI. For an AI coding workflow, pair it with an embedding model like nomic-embed-text for local RAG.
Recommended Builds
Complete PC builds that can run Qwen 2.5 Coder 32B Instruct.
AI Builder Workstation
Run every AI tool you need. Nothing leaves your machine
Runs 10 models
AMD AI Powerhouse
24 GB of AI power at nearly half the NVIDIA price
Runs 7 models
Extreme AI Workstation
Dual GPUs that run the biggest AI models at a smart price
Runs 8 models
High-End AI Workstation
Chat, generate images, and code with AI, all at once
Runs 8 models
High-End Home AI Server
Your household's private AI: chatbots, code tools, and more
Runs 12 models
Mac Studio AI Builder
Plug in and run AI: silent, powerful, no assembly required
Runs 6 models
Mid-Range AI Workstation
The sweet spot for AI: handles most models without overspending
Runs 8 models
Mid-Range Home AI Server
Serve multiple AI models to every device at home
Runs 9 models
Next-Gen AI Workstation
The fastest single GPU, built around the new RTX 5090
Runs 6 models
Silent Mini-ITX AI Box
Whisper-quiet AI processing for noise-sensitive environments
Runs 8 models
Frequently Asked Questions
- How much VRAM does Qwen 2.5 Coder 32B Instruct need?
- Qwen 2.5 Coder 32B Instruct requires 21.9 GB VRAM at recommended quality (Q5_K_M). At lower quality settings, it can fit in as little as 11.6 GB.
- What is the best GPU for Qwen 2.5 Coder 32B Instruct?
- The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Qwen 2.5 Coder 32B Instruct, achieving 140 tok/s at Q5_K_M with an excellent rating.
- Can I run Qwen 2.5 Coder 32B Instruct on an RTX 4060 Ti?
- Yes. On the NVIDIA GeForce RTX 4060 Ti 16GB, Qwen 2.5 Coder 32B Instruct runs at 10 tok/s (Q3_K_M, acceptable).
- What quantization should I use for Qwen 2.5 Coder 32B Instruct?
- For the best quality, use Q5_K_M (21.9 GB VRAM). If your GPU has limited VRAM, Q2_K (11.6 GB) is the most efficient option with acceptable quality.
- Is Qwen 2.5 Coder 32B Instruct good for coding?
- Yes. Qwen 2.5 Coder 32B Instruct is used with Cursor, Continue, Aider, Windsurf, Codex CLI for local AI coding. For the best coding experience, pair it with an embedding model for local RAG.
Related Guides
Data confidence: verified. Source
VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Qwen is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.