RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Can I run Olmo 3 32B Think locally?

Olmo 3 32B Think by allenai has 32B parameters. At 4-bit with an 8K context it needs about 20.0 GB of GPU memory. It fits on 48 of 65 GPUs and Macs we track. See all models that fit in 24 GB of VRAM.

20.0 GBQ4_K_M · 8K context
RTX 3090smallest consumer GPU that fits
M3 Pro 36GBsmallest Mac that fits
64Kmax context
GQA + sliding windowattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from bartowski/allenai_Olmo-3-32B-Think-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx
2-bit (Q2_K) •rough12.8 GB13.0 GB14.5 GB
3-bit (Q3_K_M) •usable16.1 GB16.4 GB17.9 GB
4-bit (Q4_K_M) •good, most popular19.7 GB20.0 GB21.5 GB
5-bit (Q5_K_M) •very good22.9 GB23.1 GB24.6 GB
6-bit (Q6_K) •near lossless26.2 GB26.5 GB28.0 GB
8-bit (Q8_0) •effectively lossless33.5 GB33.7 GB35.2 GB
16-bit (F16/BF16) •original weights61.6 GB61.9 GB63.4 GB

Which GPUs and Macs run Olmo 3 32B Think?

Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).

HardwareMemoryFits?Best quantNeedsMax tok/s (est.)Details
RTX 3060 12GB12 GBNo———details
RTX 4060 Ti 16GB16 GBFitsQ2_K13.0 GB24details
RTX 309024 GBFitsQ4_K_M20.0 GB48details
RTX 409024 GBFitsQ4_K_M20.0 GB52details
RTX 509032 GBFitsQ6_K26.5 GB68details
RTX 5070 Ti16 GBFitsQ2_K13.0 GB75details
RX 7900 XTX24 GBFitsQ4_K_M20.0 GB49details
Mac M4 Pro 48GB48 GBFitsQ6_K26.5 GB10details
Mac M4 Max 128GB128 GBFitsF1661.9 GB8.5details
Mac M3 Ultra 512GB512 GBFitsF1661.9 GB13details
DGX Spark128 GBFitsF1661.9 GB4.2details
RTX 407012 GBNo———details
RTX 5060 Ti 16GB16 GBFitsQ2_K13.0 GB37details
RTX 40608 GBNo———details
Arc B58012 GBNo———details
RTX 3060 Ti8 GBNo———details
RTX 30708 GBNo———details
RTX 3080 10GB10 GBNo———details
RTX 3080 12GB12 GBNo———details
RTX 3090 Ti24 GBFitsQ4_K_M20.0 GB52details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBNo———details
RTX 4070 Ti12 GBNo———details
RTX 4070 Ti Super16 GBFitsQ2_K13.0 GB56details
RTX 408016 GBFitsQ2_K13.0 GB60details
RTX 4080 Super16 GBFitsQ2_K13.0 GB61details
RTX 50608 GBNo———details
RTX 507012 GBNo———details
RTX 508016 GBFitsQ2_K13.0 GB80details
2x RTX 30902×24 GBFitsQ8_033.7 GB27details
2x RTX 40902×24 GBFitsQ8_033.7 GB29details
2x RTX 50902×32 GBFitsQ8_033.7 GB52details
RTX A600048 GBFitsQ8_033.7 GB22details
RTX 6000 Ada48 GBFitsQ8_033.7 GB28details
RTX PRO 6000 Blackwell96 GBFitsF1661.9 GB28details
L40S48 GBFitsQ8_033.7 GB25details
A100 40GB40 GBFitsQ8_033.7 GB45details
A100 80GB80 GBFitsF1661.9 GB32details
H100 PCIe80 GBFitsF1661.9 GB31details
H100 SXM80 GBFitsF1661.9 GB52details
H200141 GBFitsF1661.9 GB74details
RX 76008 GBNo———details
RX 7700 XT12 GBNo———details
RX 7800 XT16 GBFitsQ2_K13.0 GB52details
RX 7900 GRE16 GBFitsQ2_K13.0 GB48details
RX 7900 XT20 GBFitsQ3_K_M16.4 GB51details
RX 907016 GBFitsQ2_K13.0 GB53details
RX 9070 XT16 GBFitsQ2_K13.0 GB53details
Ryzen AI Max+ 395 128GB128 GBFitsF1661.9 GB4.0details
Arc A770 16GB16 GBFitsQ2_K13.0 GB47details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBFitsQ2_K13.0 GB8.3details
Mac M3 24GB24 GBFitsQ2_K13.0 GB8.3details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBFitsQ3_K_M16.4 GB7.7details
Mac M1 Pro 32GB32 GBFitsQ3_K_M16.4 GB13details
Mac M3 Pro 36GB36 GBFitsQ4_K_M20.0 GB7.7details
Mac M4 Pro 24GB24 GBFitsQ2_K13.0 GB23details
Mac M4 Pro 64GB64 GBFitsQ8_033.7 GB8.0details
Mac M1 Max 64GB64 GBFitsQ8_033.7 GB12details
Mac M2 Max 96GB96 GBFitsF1661.9 GB6.2details
Mac M3 Max 128GB128 GBFitsF1661.9 GB6.2details
Mac M4 Max 64GB64 GBFitsQ8_033.7 GB16details
Mac M2 Ultra 192GB192 GBFitsF1661.9 GB12details
Mac M3 Ultra 256GB256 GBFitsF1661.9 GB13details

Questions

How much VRAM does Olmo 3 32B Think need?

About 20.0 GB at 4-bit (Q4_K_M) with an 8K context: 18.1 GB of weights (real GGUF file size), 1.3 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 33.7 GB.

What is the smallest GPU that runs Olmo 3 32B Think?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 3090 (24 GB).

Can I run Olmo 3 32B Think on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M3 Pro 36GB or larger, using the default macOS GPU memory limit.

Similar models

Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.

Molmo2 8B

8.7B · 6.5 GB at 4-bit

olmOCR 2 7B 1025

8.3B · 5.5 GB at 4-bit

olmOCR 7B 0225 Preview

8.3B · 5.5 GB at 4-bit

Granite 4.0 H Small

32B · 18.9 GB at 4-bit

GLM Z1 32B 0414

33B · 19.5 GB at 4-bit

GLM 4 32B 0414

33B · 19.5 GB at 4-bit

Embed this

Olmo 3 32B Think: runs on 24 GB+ VRAM Free badge and "Will it run on my GPU?" widget for model cards, READMEs and blogs. No cookies, no tracking. All options.

Copy a snippet

Markdown badge

[![Olmo 3 32B Think: runs on 24 GB+ VRAM (RunsOnMy)](https://runsonmy.com/badge/olmo-3-32b-think.svg)](https://runsonmy.com/model/olmo-3-32b-think/)

Widget (script)

<div class="runsonmy-widget" data-model="olmo-3-32b-think"><a href="https://runsonmy.com/model/olmo-3-32b-think/">Will Olmo 3 32B Think run on my GPU? Check on RunsOnMy</a></div>
<script async src="https://runsonmy.com/embed.js"></script>

Widget (iframe)

<iframe src="https://runsonmy.com/embed/m/olmo-3-32b-think/" width="100%" height="220" style="max-width:440px;border:0" loading="lazy" title="Will Olmo 3 32B Think run on my GPU? (RunsOnMy)"></iframe>

Model facts

Hugging Face repoallenai/Olmo-3-32B-Think
Parameters32,233,522,176 (32B)
Layers / hidden size64 / 5120
AttentionGQA + sliding window; 40 heads, 8 KV heads, head dim 128; sliding window 4096
KV cache per token (FP16)256.0 KiB (sliding layers capped at the window)
Max context65,536 tokens
Licenseapache-2.0
Released on HFNov 19, 2025
Downloads (30d) / likes11,755 / 177

Sources: Hugging Face model API · config.json (allenai/Olmo-3-32B-Think) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.