Shanghai AI Lab20B~13 GB VRAM at Q4_K_M

InternLM 3 20B Instruct — VRAM, speed & local setup

Written by Jakub Rusinowski · Last updated

A powerful mid-size bilingual model. Exceptional at long-form document understanding, scientific reasoning, and code generation in both Chinese and English contexts.

InternLM 3 20B Instruct needs about 13 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.

Logic92
Creative82
Coding89

VRAM and speed by quantization

Quoted against NVIDIA RTX 4090 (24 GB). Weights plus overhead only: this model's architecture is unpublished, so no KV cache is added.

QuantVRAMSpeed (est.)Fit
Q2_K
2.63 bpw
7.4 GB~83 tok/sFits
Q3_K_M
3.41 bpw
9.3 GB~69 tok/sFits
Q4_K_M
4.83 bpw
12.9 GB~52 tok/sFits
Q5_K_M
5.67 bpw
15 GB~45 tok/sFits
Q6_K
6.56 bpw
17.2 GB~40 tok/sFits
Q8_0
8.50 bpw
22.1 GB~32 tok/sTight
F16
16.00 bpw
40.8 GB~3 tok/sOffload

Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. InternLM 3 20B Instruct VRAM calculator →

Get InternLM 3 20B Instruct running

The cheapest catalogued GPU that runs InternLM 3 20B Instruct is the AMD Radeon RX 9060 XT 16GB (16 GB).

Buy this hardware AMD Radeon RX 9060 XT 16GB — 16 GB VRAM · 160 W board powerDeploy in the cloud now RTX 4090 on RunPod — from $0.34/hr · rate checked 2026-07

or compare on Vast.ai from $0.35/hr (typical low · varies)

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Affiliate disclosure: Some links on this page are affiliate links — if you buy through them, LLM Configurator may earn a commission at no extra cost to you. As an Amazon Associate, LLM Configurator earns from qualifying purchases.
AMD Radeon RX 9060 XT 16GB
16 GB VRAM · 160 W board power
2026 prices are volatile — check the current listing.

How to run InternLM 3 20B Instruct

Install Ollama, then run:

ollama run internlm3:20b
Weights on Hugging Face: internlm/internlm3-20b-instruct ↗

Specifications

Curated — A hand-written entry from before this catalogue recorded its sources. The figures are long-standing but their provenance is not on file.

Parameters
20 Billion
Context window
32,768
Architecture
Dense
Provider
Shanghai AI Lab
Licence
Apache 2.0
Practical deployment
Laptop
Specified at
Q4_K_M
System RAM
32 GB
Record updated
2025-01-15
LicenceApache-2.0Commercial use permitted

Commercial use permitted. No usage restrictions beyond attribution.

Quality and use cases

Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.

Best forbilingualdocument analysisresearchcomplex reasoning

Can I run InternLM 3 20B Instruct on my GPU?

Other InternLM 3 sizes

InternLM 3 20B Instruct — frequently asked questions

How much VRAM does InternLM 3 20B Instruct need?

About 13 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.

Does InternLM 3 20B Instruct run on an RTX 4090 (24 GB)?

Yes. InternLM 3 20B Instruct needs about 13 GB at Q4_K_M, inside a 24 GB card, at an estimated 52 tokens/sec.

How do I run InternLM 3 20B Instruct locally?

Install Ollama and run `ollama run internlm3:20b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does InternLM 3 come in?

InternLM 3 8B Instruct (6 GB), InternLM 3 20B Instruct (13 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.