RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run Qwen3.5 0.8B locally?

Qwen3.5 0.8B by Qwen has 873M parameters. At 4-bit with an 8K context it needs about 1.5 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.

1.5 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest 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 unsloth/Qwen3.5-0.8B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K)rough1.2 GB1.3 GB1.6 GB2.7 GB
3-bit (Q3_K_M) •usable1.4 GB1.4 GB1.7 GB2.8 GB
4-bit (Q4_K_M) •good, most popular1.4 GB1.5 GB1.7 GB2.9 GB
5-bit (Q5_K_M) •very good1.5 GB1.5 GB1.8 GB2.9 GB
6-bit (Q6_K) •near lossless1.5 GB1.6 GB1.8 GB3.0 GB
8-bit (Q8_0) •effectively lossless1.7 GB1.7 GB2.0 GB3.1 GB
16-bit (F16/BF16) •original weights2.3 GB2.4 GB2.7 GB3.8 GB

Which GPUs and Macs run Qwen3.5 0.8B?

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.)
RTX 3060 12GB12 GBFitsF162.4 GB237details
RTX 4060 Ti 16GB16 GBFitsF162.4 GB190details
RTX 309024 GBFitsF162.4 GB617details
RTX 409024 GBFitsF162.4 GB665details
RTX 509032 GBFitsF162.4 GB1181details
RTX 5070 Ti16 GBFitsF162.4 GB591details
RX 7900 XTX24 GBFitsF162.4 GB633details
Mac M4 Pro 48GB48 GBFitsF162.4 GB180details
Mac M4 Max 128GB128 GBFitsF162.4 GB360details
Mac M3 Ultra 512GB512 GBFitsF162.4 GB540details
DGX Spark128 GBFitsF162.4 GB180details
RTX 407012 GBFitsF162.4 GB332details
RTX 5060 Ti 16GB16 GBFitsF162.4 GB295details
RTX 40608 GBFitsF162.4 GB179details
Arc B58012 GBFitsF162.4 GB301details
RTX 3060 Ti8 GBFitsF162.4 GB295details
RTX 30708 GBFitsF162.4 GB295details
RTX 3080 10GB10 GBFitsF162.4 GB501details
RTX 3080 12GB12 GBFitsF162.4 GB601details
RTX 3090 Ti24 GBFitsF162.4 GB665details
RTX 4060 Ti 8GB8 GBFitsF162.4 GB190details
RTX 4070 Super12 GBFitsF162.4 GB332details
RTX 4070 Ti12 GBFitsF162.4 GB332details
RTX 4070 Ti Super16 GBFitsF162.4 GB443details
RTX 408016 GBFitsF162.4 GB473details
RTX 4080 Super16 GBFitsF162.4 GB485details
RTX 50608 GBFitsF162.4 GB295details
RTX 507012 GBFitsF162.4 GB443details
RTX 508016 GBFitsF162.4 GB633details
2x RTX 30902×24 GBFitsF162.4 GB617details
2x RTX 40902×24 GBFitsF162.4 GB665details
2x RTX 50902×32 GBFitsF162.4 GB1181details
RTX A600048 GBFitsF162.4 GB506details
RTX 6000 Ada48 GBFitsF162.4 GB633details
RTX PRO 6000 Blackwell96 GBFitsF162.4 GB1181details
L40S48 GBFitsF162.4 GB570details
A100 40GB40 GBFitsF162.4 GB1025details
A100 80GB80 GBFitsF162.4 GB1344details
H100 PCIe80 GBFitsF162.4 GB1319details
H100 SXM80 GBFitsF162.4 GB2209details
H200141 GBFitsF162.4 GB3165details
RX 76008 GBFitsF162.4 GB190details
RX 7700 XT12 GBFitsF162.4 GB285details
RX 7800 XT16 GBFitsF162.4 GB411details
RX 7900 GRE16 GBFitsF162.4 GB380details
RX 7900 XT20 GBFitsF162.4 GB527details
RX 907016 GBFitsF162.4 GB422details
RX 9070 XT16 GBFitsF162.4 GB422details
Ryzen AI Max+ 395 128GB128 GBFitsF162.4 GB169details
Arc A770 16GB16 GBFitsF162.4 GB369details
Mac M1 16GB16 GBFitsF162.4 GB45details
Mac M2 24GB24 GBFitsF162.4 GB66details
Mac M3 24GB24 GBFitsF162.4 GB66details
Mac M4 16GB16 GBFitsF162.4 GB79details
Mac M4 32GB32 GBFitsF162.4 GB79details
Mac M1 Pro 32GB32 GBFitsF162.4 GB132details
Mac M3 Pro 36GB36 GBFitsF162.4 GB99details
Mac M4 Pro 24GB24 GBFitsF162.4 GB180details
Mac M4 Pro 64GB64 GBFitsF162.4 GB180details
Mac M1 Max 64GB64 GBFitsF162.4 GB264details
Mac M2 Max 96GB96 GBFitsF162.4 GB264details
Mac M3 Max 128GB128 GBFitsF162.4 GB264details
Mac M4 Max 64GB64 GBFitsF162.4 GB360details
Mac M2 Ultra 192GB192 GBFitsF162.4 GB527details
Mac M3 Ultra 256GB256 GBFitsF162.4 GB540details

Questions

How much VRAM does Qwen3.5 0.8B need?

About 1.5 GB at 4-bit (Q4_K_M) with an 8K context: 0.5 GB of weights (real GGUF file size), 0.1 GB of KV cache and 0.9 GB of runtime overhead. At 8-bit it needs about 1.7 GB.

What is the smallest GPU that runs Qwen3.5 0.8B?

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 Qwen3.5 0.8B 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.

Qwen3 0.6B

752M · 1.9 GB at 4-bit

Qwen2.5 0.5B Instruct

494M · 1.2 GB at 4-bit

Qwen2.5 1.5B Instruct

1.5B · 1.9 GB at 4-bit

LightOnOCR 3 0.8B

853M · 1.5 GB at 4-bit

PaddleOCR VL 1.6

959M · 1.3 GB at 4-bit

FastVLM 0.5B

759M · 1.2 GB at 4-bit

Model facts

Hugging Face repoQwen/Qwen3.5-0.8B
Parameters873,438,784 (873M)
Layers / hidden size24 / 1024
AttentionGQA (hybrid: 18 linear/state layers); 8 heads, 2 KV heads, head dim 256
KV cache per token (FP16)12.0 KiB
Max context262,144 tokens
Licenseapache-2.0
Released on HFFeb 28, 2026
Downloads (30d) / likes2,900,659 / 847

Sources: Hugging Face model API · config.json (Qwen/Qwen3.5-0.8B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.