Skip to content
localmodel.run

Text model · RNJ

RN RNJ-1 8B: RAM and VRAM requirements

RNJ-1 8B needs about 6.3 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~4.76 GB to download; KV cache and overhead add the rest), or about 9.7 GB at Q8_0. The lightest hardware that runs it is Nvidia GeForce RTX 3060 Ti (8GB).

RNJ family · 8B params · released Dec 2025 · 504.9K Ollama pulls.

Shopping for hardware? See what runs RNJ-1 8B →

License Apache-2.0 · Commercial OK ↓ 2.4K/mo ♥ 321 on HuggingFace
Q4_K_M GGUF
4.76 GB
Q8_0 GGUF
8.23 GB
Memory @ Q4 (4k)
~6.3 GB
Context
32 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

RNJ-1 8B runs on 35 of 43 tracked devices at Q4_K_M.

33 run well 2 tight fit 8 too small
Fit check Q4_K_M
Yes, it runs fast
needs 6.3 GB usable 10.5 GB
See the full breakdown
$ollama run rnj-1:8b

Memory breakdown

Weights (Q4_K_M)4.76 GB
+
KV cache (4k)0.7 GB
+
Overhead0.8 GB
=
Total6.3 GB

How context length changes it

4k context ~6.3 GB 32k context ~11 GB 128k context ~27.3 GB

Longer context grows the KV cache quickly: RNJ-1 8B needs ~6.3 GB at 4k but ~27.3 GB at 128k, which can push it past a device that fits it at a short context.

Speed drops too: every token re-reads the KV cache, so at 128k context RNJ-1 8B generates at roughly ~18% of its short-context speed. How this is estimated.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 3.4 GB est
Q3_K_M 3.9 GB est
Q4_K_M (default) 4.76 GB
Q5_K_M 5.7 GB est
Q6_K 6.6 GB est
Q8_0 8.23 GB
FP16 16.63 GB

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run rnj-1:8b
llama.cpp
$ llama-cli -hf unsloth/rnj-1-instruct-GGUF:Q4_K_M
LM Studio
$ lms get unsloth/rnj-1-instruct-GGUF

Which devices can run RNJ-1 8B?

NVIDIA GPUs

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down Olmo 3 7B Instruct 7B Next size up DeepSeek-R1-0528-Qwen3-8B 8.19B

Similar models

FAQ

How much VRAM or RAM does RNJ-1 8B need?

At Q4_K_M, RNJ-1 8B needs about 6.3 GB (weights ~4.76 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~9.7 GB.

What is the Q4_K_M GGUF file size of RNJ-1 8B?

The Q4_K_M GGUF file is about 4.76 GB to download, and the Q8_0 GGUF is about 8.23 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~6.3 GB of memory at Q4_K_M.

Can RNJ-1 8B run on a laptop?

Yes, RNJ-1 8B fits on a 16 GB machine at Q4_K_M and runs on Apple Silicon or a 12 GB+ GPU comfortably.

Can I use RNJ-1 8B commercially?

Yes. RNJ-1 8B is licensed Apache-2.0, which permits commercial use.

Understand the numbers

Short guides to the ideas behind RNJ-1 8B's memory and quant figures.

Essential AI's from-scratch dense 8B, optimized for code and STEM, 32K context, Apache-2.0. Q4_K_M from unsloth matches Essential's own GGUF byte for byte (5.11GB).

Sources

Last validated 2026-09-07. Memory figures are estimates. See methodology.