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

Can I run Sage A1.5 locally?

Sage A1.5 by InternScience has 36B parameters (4.7B active per token, 256 experts, 8 used per token). At 4-bit with an 8K context it needs about 21.3 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.

21.3 GBQ4_K_M · 8K context
RTX 5090smallest consumer GPU that fits
M3 Pro 36GBsmallest Mac that fits
256Kmax context
GQAattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from mradermacher/Sage-A1.5-i1-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough13.3 GB13.4 GB13.8 GB15.7 GB
3-bit (Q3_K_M) •usable16.9 GB17.0 GB17.5 GB19.4 GB
4-bit (Q4_K_M) •good, most popular21.2 GB21.3 GB21.7 GB23.6 GB
5-bit (Q5_K_M) •very good24.6 GB24.6 GB25.1 GB27.0 GB
6-bit (Q6_K) •near lossless28.2 GB28.2 GB28.7 GB30.6 GB
8-bit (Q8_0)effectively lossless36.5 GB36.6 GB37.1 GB38.9 GB
16-bit (F16/BF16)original weights67.9 GB68.0 GB68.5 GB70.3 GB

Which GPUs and Macs run Sage A1.5?

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_K13.4 GB165details
RTX 309024 GBFitsQ3_K_M17.0 GB413details
RTX 409024 GBFitsQ3_K_M17.0 GB445details
RTX 509032 GBFitsQ5_K_M24.6 GB536details
RTX 5070 Ti16 GBFitsQ2_K13.4 GB512details
RX 7900 XTX24 GBFitsQ3_K_M17.0 GB424details
Mac M4 Pro 48GB48 GBFitsQ6_K28.2 GB71details
Mac M4 Max 128GB128 GBFitsF1668.0 GB58details
Mac M3 Ultra 512GB512 GBFitsF1668.0 GB86details
DGX Spark128 GBFitsF1668.0 GB29details
RTX 407012 GBNo———details
RTX 5060 Ti 16GB16 GBFitsQ2_K13.4 GB256details
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 GBFitsQ3_K_M17.0 GB445details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBNo———details
RTX 4070 Ti12 GBNo———details
RTX 4070 Ti Super16 GBFitsQ2_K13.4 GB384details
RTX 408016 GBFitsQ2_K13.4 GB410details
RTX 4080 Super16 GBFitsQ2_K13.4 GB421details
RTX 50608 GBNo———details
RTX 507012 GBNo———details
RTX 508016 GBFitsQ2_K13.4 GB549details
2x RTX 30902×24 GBFitsQ8_036.6 GB186details
2x RTX 40902×24 GBFitsQ8_036.6 GB200details
2x RTX 50902×32 GBFitsQ8_036.6 GB355details
RTX A600048 GBFitsQ8_036.6 GB152details
RTX 6000 Ada48 GBFitsQ8_036.6 GB190details
RTX PRO 6000 Blackwell96 GBFitsF1668.0 GB189details
L40S48 GBFitsQ8_036.6 GB171details
A100 40GB40 GBFitsQ6_K28.2 GB403details
A100 80GB80 GBFitsF1668.0 GB215details
H100 PCIe80 GBFitsF1668.0 GB211details
H100 SXM80 GBFitsF1668.0 GB353details
H200141 GBFitsF1668.0 GB506details
RX 76008 GBNo———details
RX 7700 XT12 GBNo———details
RX 7800 XT16 GBFitsQ2_K13.4 GB357details
RX 7900 GRE16 GBFitsQ2_K13.4 GB329details
RX 7900 XT20 GBFitsQ3_K_M17.0 GB353details
RX 907016 GBFitsQ2_K13.4 GB366details
RX 9070 XT16 GBFitsQ2_K13.4 GB366details
Ryzen AI Max+ 395 128GB128 GBFitsF1668.0 GB27details
Arc A770 16GB16 GBFitsQ2_K13.4 GB320details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBFitsQ2_K13.4 GB57details
Mac M3 24GB24 GBFitsQ2_K13.4 GB57details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBFitsQ3_K_M17.0 GB53details
Mac M1 Pro 32GB32 GBFitsQ3_K_M17.0 GB88details
Mac M3 Pro 36GB36 GBFitsQ4_K_M21.3 GB52details
Mac M4 Pro 24GB24 GBFitsQ2_K13.4 GB156details
Mac M4 Pro 64GB64 GBFitsQ8_036.6 GB54details
Mac M1 Max 64GB64 GBFitsQ8_036.6 GB79details
Mac M2 Max 96GB96 GBFitsQ8_036.6 GB79details
Mac M3 Max 128GB128 GBFitsF1668.0 GB42details
Mac M4 Max 64GB64 GBFitsQ8_036.6 GB108details
Mac M2 Ultra 192GB192 GBFitsF1668.0 GB84details
Mac M3 Ultra 256GB256 GBFitsF1668.0 GB86details

Questions

How much VRAM does Sage A1.5 need?

About 21.3 GB at 4-bit (Q4_K_M) with an 8K context: 20.2 GB of weights (real GGUF file size), 0.2 GB of KV cache and 0.9 GB of runtime overhead. At 8-bit it needs about 36.6 GB.

What is the smallest GPU that runs Sage A1.5?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 5090 (32 GB).

Can I run Sage A1.5 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.

Qwen3.6 35B A3B

36B · 21.8 GB at 4-bit

Ornith 1.5 35B A3B

36B · 21.4 GB at 4-bit

Qwen3.5 35B A3B

36B · 21.8 GB at 4-bit

Hermes 4.3 36B

36B · 23.0 GB at 4-bit

Seed OSS 36B Instruct

36B · 23.0 GB at 4-bit

Maxwell V1 Flash

37B · 22.1 GB at 4-bit

Embed this

Sage A1.5: runs on 32 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

[![Sage A1.5: runs on 32 GB+ VRAM (RunsOnMy)](https://runsonmy.com/badge/sage-a1-5.svg)](https://runsonmy.com/model/sage-a1-5/)

Widget (script)

<div class="runsonmy-widget" data-model="sage-a1-5"><a href="https://runsonmy.com/model/sage-a1-5/">Will Sage A1.5 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/sage-a1-5/" width="100%" height="220" style="max-width:440px;border:0" loading="lazy" title="Will Sage A1.5 run on my GPU? (RunsOnMy)"></iframe>

Model facts

Hugging Face repoInternScience/Sage-A1.5
Parameters35,951,822,704 (36B) · ~4.7B active (computed from config)
Layers / hidden size40 / 2048
AttentionGQA (hybrid: 30 linear/state layers); 16 heads, 2 KV heads, head dim 256
KV cache per token (FP16)20.0 KiB
Max context262,144 tokens
Licenseapache-2.0
Released on HFOct 15, 2026
Downloads (30d) / likes25,027 / 409

Sources: Hugging Face model API · config.json (InternScience/Sage-A1.5) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 18, 2026. Spot an error? Report it.