VOOZH about

URL: https://willitrunai.com/can-run/hf-bartowski--internlm2-5-20b-chat-gguf-on-l20-48gb

⇱ internlm2 5 20b chat on NVIDIA L20 48GB? YES

Can internlm2 5 20b chat run on NVIDIA L20 48GB?

YES — Runs Great

C49Usable
○Estimated from fit model

internlm2 5 20b chat needs ~20.5 GB VRAM. NVIDIA L20 48GB has 48.0 GB. With Q4_K_M quantization, expect ~52 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: HighStack: BasicBottleneck: Balanced
Share:

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 20.5 GB, 51.7 tok/s, Runs well
20.5 GB required48.0 GB available
43% VRAM used

Fit status

Runs well

Decode

51.7 tok/s

TTFT

3745 ms

Safe context

203K

Memory

20.5 GB / 48.0 GB

Memory breakdown

Weights12.2 GB
KV Cache2.3 GB
Runtime1.2 GB
Headroom4.8 GB

See how fast it feels

See how fast it feelsinternlm2 5 20b chat on NVIDIA L20 48GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API. When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed. Key benefits of running locally: - Full privacy: your prompts never leave your machine - No per-token cost or rate limits - Works offline once the model is downloaded - Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb: Model size Q4 (4-bit) Q8 (8-bit) FP16 7B params ~4.3 GB ~7.5 GB ~14 GB 13B params ~7.9 GB ~13.9 GB ~26 GB 70B params ~42.7 GB ~74.9 GB ~140 GB Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 51.7 tok/s decode · 3.7s TTFT (warm) · 129 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best improvement path

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatCRuns well51.7 tok/s2043 ms203K
CodingCRuns well51.7 tok/s3745 ms203K
Agentic CodingCRuns well51.7 tok/s5447 ms203K
ReasoningCRuns well51.7 tok/s4426 ms203K
RAGCRuns well51.7 tok/s6809 ms203K

Quantization options

How internlm2 5 20b chat (20B params) fits at each quantization level on NVIDIA L20 48GB (48.0 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
7.8 GB
LowC42
Q3_K_S
3
9.8 GB
LowC42
NVFP4
4
11.2 GB
MediumC43
Q4_K_M
4
12.2 GB
MediumC43
Q5_K_M
5
14.4 GB
HighC44
Q6_K
6
16.4 GB
HighC44
Q8_0
8
21.4 GB
Very HighC46
F16Best for your GPU
16
41.0 GB
MaximumC47

Get started

Copy-paste commands to run internlm2 5 20b chat on your machine.

Run

lms load hf-bartowski--internlm2-5-20b-chat-gguf && lms server start

Upgrade options

Hardware that runs internlm2 5 20b chat well

AMD Instinct MI210 64GBBudget pick
64 GB VRAM (+16)1638 GB/s (+774)
C
Raises estimated decode speed by about 77%.91.3 tok/s decode

Raises estimated decode speed by about 77%.

Adds memory headroom for longer context windows and future model growth.

~$10,000 MSRP

Frequently asked questions

See all results for NVIDIA L20 48GBSee all hardware for internlm2 5 20b chat