codium bd coder v2 lora needs about 10.0 GB of memory at its main file (a 8.4 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 4.4 GB.
Codium BD Coder v2 – Bengali‑sourced coding and vision AI
Codium BD Coder v2 is a 4.21 billion‑parameter Qwen‑35 model optimized with 4‑bit QLoRA. It supports coding, reasoning, tools, and vision tasks. The model is released under the Apache‑2.0 license.
Good at
- Bengali and English code generation
- Multimodal vision analysis
- Tool‑aware instruction following
Limits
- No vision support on CPU‑only setups
- Requires large VRAM for full context
- License allows commercial use but no redistribution of weights
Best for
- Developers needing Bengali code assistance
- Rapid prototyping of UI/UX designs
- Automated code review and debugging
Who it’s for. This model is suitable for software engineers, educators, and researchers who need a lightweight, multilingual coding assistant with vision capabilities. It is ideal for local deployments on consumer GPUs with sufficient VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 8.4 GB | 10.0 GB | Too big | Get file → |
The available GGUF file is 8.42 GB and estimates 10 GB of VRAM at 8 k context. For 8 GB VRAM, the model will not fit; consider using a smaller context or a different quantization if available. If you have 10 GB or more VRAM, the current quantization will run with moderate performance.
What GPU do you need for codium bd coder v2 lora?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is the main file (8.4 GB file, about 10.0 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about codium bd coder v2 lora
How much VRAM does codium bd coder v2 lora need?
About 10.0 GB at its main file with an 8K context window. That is the 8.4 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 codium bd coder v2 lora on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of codium bd coder v2 lora GGUF?
It depends on the quant: file is 8.4 GB.
What context length does codium bd coder v2 lora 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 codium bd coder v2 lora GGUF?
The files are on Hugging Face in the ganjakha/codium-bd-coder-v2-lora repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does codium bd coder v2 lora 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 Codium BD Coder v2 on 8 GB VRAM?
The current GGUF file requires approximately 10 GB of VRAM at 8 k context, so it will not run on 8 GB VRAM.
Does the model support vision tasks?
Yes, it has multimodal vision support for analyzing system architectures and UI screenshots.
What license does the model use?
The model is released under the Apache‑2.0 license.