CaPtain 4B needs about 4.7 GB of memory at Q4_K_M (a 3.1 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 12 quants, from 2.1 GB (Q2_K) to 9.7 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.4 GB.
CaPtain 4B – a Qwen3.5‑based code‑writing agent
CaPtain 4B is a 4.84 billion‑parameter model fine‑tuned for code‑as‑policy robot control. It generates Python for the CaPtain‑Harness and interacts with vision services. The model is released under apache‑2.0.
Good at
- Python code generation for robotics
- Tool‑based perception integration
- Reasoning with large context
Limits
- No direct vision output
- Requires CaPtain‑Harness to run
- No support for non‑Python tools
Best for
- Robot control via code writing
- Simulated robotics experiments
- Tool‑driven AI pipelines
Who it’s for. Developers building robot agents that need a code‑writing policy. Researchers testing reinforcement learning with code‑as‑policy agents. Engineers integrating vision services into a Python harness.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 2.1 GB | 3.7 GB | Fits | Get file → | |
| Q3_K_S | 2.3 GB | 4.0 GB | Fits | Get file → | |
| Q3_K_M | 2.5 GB | 4.1 GB | Fits | Get file → | |
| Q3_K_L | 2.7 GB | 4.3 GB | Fits | Get file → | |
| IQ4_XS | 2.9 GB | 4.5 GB | Fits | Get file → | |
| Q4_K_S | 2.9 GB | 4.5 GB | Fits | Get file → | |
| Q4_K_M | 3.1 GB | 4.7 GB | Fits | Get file → | |
| Q5_K_S | 3.4 GB | 5.0 GB | Fits | Get file → | |
| Q5_K_M | 3.5 GB | 5.1 GB | Fits | Get file → | |
| Q6_K | 4.0 GB | 5.6 GB | Fits | Get file → | |
| Q8_0 | 5.2 GB | 6.8 GB | Fits | Get file → | |
| F16 | 9.7 GB | 11.3 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (3.07 GB, 4.7 GB memory). For 12 GB VRAM, Q5_K_M (3.51 GB, 5.1 GB memory). For 16 GB VRAM, Q6_K (3.99 GB, 5.6 GB memory). Estimates are approximate.
What GPU do you need for CaPtain 4B?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q3_K_S (2.3 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K (4.0 GB file, about 5.6 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (5.2 GB file, about 6.8 GB in use).
- 12 GB VRAM: the best quant that fits is F16 (9.7 GB file, about 11.3 GB in use).
Which CaPtain 4B quant should you choose?
Q4_K_M is the usual starting point for CaPtain 4B: it balances file size and quality well. Here it is a 3.1 GB file.
The Q8_0 file is 5.2 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 2.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about CaPtain 4B
How much VRAM does CaPtain 4B need?
About 4.7 GB at Q4_K_M with an 8K context window. That is the 3.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run CaPtain 4B on an 8 GB GPU?
Yes, with Q8_0: the 5.2 GB file needs about 6.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of CaPtain 4B GGUF?
It depends on the quant: Q2_K is 2.1 GB, Q3_K_S is 2.3 GB, Q3_K_M is 2.5 GB, Q3_K_L is 2.7 GB, IQ4_XS is 2.9 GB, Q4_K_S is 2.9 GB and 6 more.
What context length does CaPtain 4B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.4 GB at 32K.
Where can I download CaPtain 4B GGUF?
The files are on Hugging Face in the mradermacher/CaPtain-4B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does CaPtain 4B use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run CaPtain 4B on 8 GB VRAM?
Yes, the Q4_K_M quant uses about 4.7 GB memory at 8 k context, which fits 8 GB VRAM with headroom.
Does CaPtain 4B support vision directly?
No, it calls external vision tools; the model itself does not output vision data.