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

Can I run Tiny Aya Global locally?

Tiny Aya Global by CohereLabs has 3.3B parameters. At 4-bit with an 8K context it needs about 3.2 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track. See all models that fit in 8 GB of VRAM.

3.2 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest Mac that fits
488Kmax 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 CohereLabs/tiny-aya-global-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K)rough2.4 GB2.4 GB2.8 GB4.5 GB
3-bit (Q3_K_M)usable2.7 GB2.8 GB3.2 GB4.9 GB
4-bit (Q4_K_M) •good, most popular3.2 GB3.2 GB3.7 GB5.4 GB
5-bit (Q5_K_M)very good3.4 GB3.5 GB3.9 GB5.6 GB
6-bit (Q6_K)near lossless3.7 GB3.8 GB4.2 GB5.9 GB
8-bit (Q8_0) •effectively lossless4.5 GB4.6 GB5.0 GB6.7 GB
16-bit (F16/BF16) •original weights7.4 GB7.5 GB7.9 GB9.6 GB

Which GPUs and Macs run Tiny Aya Global?

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 GBFitsF167.5 GB54details
RTX 4060 Ti 16GB16 GBFitsF167.5 GB43details
RTX 309024 GBFitsF167.5 GB139details
RTX 409024 GBFitsF167.5 GB150details
RTX 509032 GBFitsF167.5 GB267details
RTX 5070 Ti16 GBFitsF167.5 GB134details
RX 7900 XTX24 GBFitsF167.5 GB143details
Mac M4 Pro 48GB48 GBFitsF167.5 GB41details
Mac M4 Max 128GB128 GBFitsF167.5 GB81details
Mac M3 Ultra 512GB512 GBFitsF167.5 GB122details
DGX Spark128 GBFitsF167.5 GB41details
RTX 407012 GBFitsF167.5 GB75details
RTX 5060 Ti 16GB16 GBFitsF167.5 GB67details
RTX 40608 GBFitsQ8_04.6 GB76details
Arc B58012 GBFitsF167.5 GB68details
RTX 3060 Ti8 GBFitsQ8_04.6 GB125details
RTX 30708 GBFitsQ8_04.6 GB125details
RTX 3080 10GB10 GBFitsF167.5 GB113details
RTX 3080 12GB12 GBFitsF167.5 GB136details
RTX 3090 Ti24 GBFitsF167.5 GB150details
RTX 4060 Ti 8GB8 GBFitsQ8_04.6 GB81details
RTX 4070 Super12 GBFitsF167.5 GB75details
RTX 4070 Ti12 GBFitsF167.5 GB75details
RTX 4070 Ti Super16 GBFitsF167.5 GB100details
RTX 408016 GBFitsF167.5 GB107details
RTX 4080 Super16 GBFitsF167.5 GB110details
RTX 50608 GBFitsQ8_04.6 GB125details
RTX 507012 GBFitsF167.5 GB100details
RTX 508016 GBFitsF167.5 GB143details
2x RTX 30902×24 GBFitsF167.5 GB139details
2x RTX 40902×24 GBFitsF167.5 GB150details
2x RTX 50902×32 GBFitsF167.5 GB267details
RTX A600048 GBFitsF167.5 GB114details
RTX 6000 Ada48 GBFitsF167.5 GB143details
RTX PRO 6000 Blackwell96 GBFitsF167.5 GB267details
L40S48 GBFitsF167.5 GB129details
A100 40GB40 GBFitsF167.5 GB232details
A100 80GB80 GBFitsF167.5 GB304details
H100 PCIe80 GBFitsF167.5 GB298details
H100 SXM80 GBFitsF167.5 GB499details
H200141 GBFitsF167.5 GB715details
RX 76008 GBFitsQ8_04.6 GB81details
RX 7700 XT12 GBFitsF167.5 GB64details
RX 7800 XT16 GBFitsF167.5 GB93details
RX 7900 GRE16 GBFitsF167.5 GB86details
RX 7900 XT20 GBFitsF167.5 GB119details
RX 907016 GBFitsF167.5 GB95details
RX 9070 XT16 GBFitsF167.5 GB95details
Ryzen AI Max+ 395 128GB128 GBFitsF167.5 GB38details
Arc A770 16GB16 GBFitsF167.5 GB83details
Mac M1 16GB16 GBFitsF167.5 GB10details
Mac M2 24GB24 GBFitsF167.5 GB15details
Mac M3 24GB24 GBFitsF167.5 GB15details
Mac M4 16GB16 GBFitsF167.5 GB18details
Mac M4 32GB32 GBFitsF167.5 GB18details
Mac M1 Pro 32GB32 GBFitsF167.5 GB30details
Mac M3 Pro 36GB36 GBFitsF167.5 GB22details
Mac M4 Pro 24GB24 GBFitsF167.5 GB41details
Mac M4 Pro 64GB64 GBFitsF167.5 GB41details
Mac M1 Max 64GB64 GBFitsF167.5 GB60details
Mac M2 Max 96GB96 GBFitsF167.5 GB60details
Mac M3 Max 128GB128 GBFitsF167.5 GB60details
Mac M4 Max 64GB64 GBFitsF167.5 GB81details
Mac M2 Ultra 192GB192 GBFitsF167.5 GB119details
Mac M3 Ultra 256GB256 GBFitsF167.5 GB122details

Questions

How much VRAM does Tiny Aya Global need?

About 3.2 GB at 4-bit (Q4_K_M) with an 8K context: 2.0 GB of weights (real GGUF file size), 0.4 GB of KV cache and 0.9 GB of runtime overhead. At 8-bit it needs about 4.6 GB.

What is the smallest GPU that runs Tiny Aya Global?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 4060 (8 GB).

Can I run Tiny Aya Global on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M1 16GB 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.

Aya Expanse 8B

8.0B · 6.6 GB at 4-bit

North Mini Code 1.0

30B · 18.8 GB at 4-bit

Unlimited OCR 2

3.3B · 3.0 GB at 4-bit

Unlimited OCR

3.3B · 2.9 GB at 4-bit

DeepSeek OCR

3.3B · 2.9 GB at 4-bit

DeepSeek VL2 Tiny

3.4B · 3.0 GB at 4-bit

Embed this

Tiny Aya Global: runs on 8 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

[![Tiny Aya Global: runs on 8 GB+ VRAM (RunsOnMy)](https://runsonmy.com/badge/tiny-aya-global.svg)](https://runsonmy.com/model/tiny-aya-global/)

Widget (script)

<div class="runsonmy-widget" data-model="tiny-aya-global"><a href="https://runsonmy.com/model/tiny-aya-global/">Will Tiny Aya Global 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/tiny-aya-global/" width="100%" height="220" style="max-width:440px;border:0" loading="lazy" title="Will Tiny Aya Global run on my GPU? (RunsOnMy)"></iframe>

Model facts

Hugging Face repoCohereLabs/tiny-aya-global
Parameters3,349,227,520 (3.3B)
Layers / hidden size36 / 2048
AttentionGQA + sliding window; 16 heads, 4 KV heads, head dim 128; sliding window 4096
KV cache per token (FP16)72.0 KiB (sliding layers capped at the window)
Max context500,000 tokens
Licensecc-by-nc-4.0 (gated: accept terms on Hugging Face)
Released on HFFeb 13, 2026
Downloads (30d) / likes31,652 / 210

Sources: Hugging Face model API · config.json (mlx-community/tiny-aya-global-8bit-mlx) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.