TR

TroyDoesAGI

TroydoesagiGGUF by mradermacherUpdated 2024-08-02
Download on Hugging Face
Parameters15.2B
Context32K
Layers38
Architecturellama
Smallest quant3.3 GB
Quick answer

TroyDoesAGI needs about 10.9 GB of memory at Q4_K_M (a 9.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 21 quants, from 3.3 GB (IQ1_S) to 12.5 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 5.1 GB.

15B LLaMA model with 32k context, optimized for context obedience

TroyDoesAGI is a 15.2B parameter LLaMA model trained for 32k context length. It focuses on adhering to provided context and reducing hallucinations, suitable for RAG workflows. The model is available in multiple GGUF quantizations.

Good at

  • Context‑obedient responses
  • Reduced hallucinations
  • RAG‑friendly prompts

Limits

  • No vision support
  • CC‑BY‑ND license restricts commercial use
  • Requires 8k context for memory estimates

Best for

  • RAG applications
  • Instruction following with context tags
  • Low‑hallucination chat

Who it’s for. Developers building local LLM services that need strict context adherence. Researchers testing context‑obeying prompts. Users who prefer a 15B model with a 32k context window.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S3.3 GB 5.1 GB
Fits Get file →
IQ1_M3.6 GB 5.4 GB
Fits Get file →
IQ2_XXS4.1 GB 5.9 GB
Fits Get file →
IQ2_XS4.6 GB 6.3 GB
Fits Get file →
IQ2_S4.8 GB 6.6 GB
Fits Get file →
IQ2_M5.2 GB 7.0 GB
Fits Get file →
Q2_K5.7 GB 7.5 GB
Tight Get file →
IQ3_XXS5.9 GB 7.7 GB
Tight Get file →
IQ3_XS6.3 GB 8.1 GB
Too big Get file →
Q3_K_S6.6 GB 8.4 GB
Too big Get file →
IQ3_S6.7 GB 8.4 GB
Too big Get file →
IQ3_M6.9 GB 8.7 GB
Too big Get file →
Q3_K_M7.4 GB 9.2 GB
Too big Get file →
Q3_K_L8.0 GB 9.8 GB
Too big Get file →
IQ4_XS8.2 GB 10.0 GB
Too big Get file →
Q4_08.6 GB 10.4 GB
Too big Get file →
Q4_K_S8.7 GB 10.4 GB
Too big Get file →
Q4_K_M9.1 GB 10.9 GB
Too big Get file →
Q5_K_S10.5 GB 12.3 GB
Too big Get file →
Q5_K_M10.8 GB 12.5 GB
Too big Get file →
Q6_K12.5 GB 14.3 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (3.33 GB, 5.1 GB memory at 8k context). If 12 GB VRAM is available, IQ2_S (4.84 GB, 6.6 GB memory) or Q4_K_M (9.12 GB, 10.9 GB memory) are options. Higher precision quants like Q5_K_M or Q6_K provide better quality at the cost of more memory.

What GPU do you need for TroyDoesAGI?

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: the best quant that fits is IQ2_XXS (4.1 GB file, about 5.9 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ3_XXS (5.9 GB file, about 7.7 GB in use).
  • 12 GB VRAM: the best quant that fits is Q4_K_M (9.1 GB file, about 10.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q6_K (12.5 GB file, about 14.3 GB in use).

Which TroyDoesAGI quant should you choose?

Q4_K_M is the usual starting point for TroyDoesAGI: it balances file size and quality well. Here it is a 9.1 GB file.

If memory is tight, the smallest file is IQ1_S at 3.3 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

Architecturellama
Parameters15.2B
Layers38
Trained context32,768 tokens
KV cache (fp16)152 KB per token

Source

Quantized bymradermacher
Downloads (HF)445
Last updated2024-08-02

Frequently asked questions about TroyDoesAGI

How much VRAM does TroyDoesAGI need?

About 10.9 GB at Q4_K_M with an 8K context window. That is the 9.1 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 TroyDoesAGI on an 8 GB GPU?

Yes, with IQ3_XXS: the 5.9 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of TroyDoesAGI GGUF?

It depends on the quant: IQ1_S is 3.3 GB, IQ1_M is 3.6 GB, IQ2_XXS is 4.1 GB, IQ2_XS is 4.6 GB, IQ2_S is 4.8 GB, IQ2_M is 5.2 GB and 15 more.

What context length does TroyDoesAGI support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.3 GB at 8K and 5.1 GB at 32K.

Where can I download TroyDoesAGI GGUF?

The files are on Hugging Face in the mradermacher/TroyDoesAGI-i1-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 TroyDoesAGI use?

Hugging Face lists the license as cc-by-nd-4.0. 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 TroyDoesAGI on 8 GB VRAM?

Yes, the IQ1_S quant uses 3.33 GB file size and estimates 5.1 GB VRAM at 8k context.

Does this model support vision?

No, the model has no vision capabilities.

What license applies to this model?

The model is licensed under cc‑by‑nd‑4.0, which allows sharing but not commercial use.

Other GGUFs of TroyDoesAGI

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Troydoesagi models

All Troydoesagi →

Other medium models

All medium models →