RunsOnMy ModelsGPUs & MacsAPI pricesVRAM calculatorNew

Can I run Yi 6B locally?

Yi 6B by 01-ai has 6.1B parameters. At 4-bit with an 8K context it needs about 4.4 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track.

4.4 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest Mac that fits
4Kmax 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/Yi-6B-i1-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx
2-bit (Q2_K) •rough2.9 GB
3-bit (Q3_K_M) •usable3.6 GB
4-bit (Q4_K_M) •good, most popular4.2 GB
5-bit (Q5_K_M) •very good4.8 GB
6-bit (Q6_K) •near lossless5.4 GB
8-bit (Q8_0)effectively lossless6.8 GB
16-bit (F16/BF16)original weights12.1 GB

Which GPUs and Macs run Yi 6B?

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 GBFitsQ8_07.0 GB56details
RTX 4060 Ti 16GB16 GBFitsF1612.3 GB24details
RTX 309024 GBFitsF1612.3 GB77details
RTX 409024 GBFitsF1612.3 GB83details
RTX 509032 GBFitsF1612.3 GB148details
RTX 5070 Ti16 GBFitsF1612.3 GB74details
RX 7900 XTX24 GBFitsF1612.3 GB79details
Mac M4 Pro 48GB48 GBFitsF1612.3 GB23details
Mac M4 Max 128GB128 GBFitsF1612.3 GB45details
Mac M3 Ultra 512GB512 GBFitsF1612.3 GB68details
DGX Spark128 GBFitsF1612.3 GB23details
RTX 407012 GBFitsQ8_07.0 GB78details
RTX 5060 Ti 16GB16 GBFitsF1612.3 GB37details
RTX 40608 GBFitsQ6_K5.7 GB55details
Arc B58012 GBFitsQ8_07.0 GB71details
RTX 3060 Ti8 GBFitsQ6_K5.7 GB90details
RTX 30708 GBFitsQ6_K5.7 GB90details
RTX 3080 10GB10 GBFitsQ8_07.0 GB118details
RTX 3080 12GB12 GBFitsQ8_07.0 GB142details
RTX 3090 Ti24 GBFitsF1612.3 GB83details
RTX 4060 Ti 8GB8 GBFitsQ6_K5.7 GB58details
RTX 4070 Super12 GBFitsQ8_07.0 GB78details
RTX 4070 Ti12 GBFitsQ8_07.0 GB78details
RTX 4070 Ti Super16 GBFitsF1612.3 GB55details
RTX 408016 GBFitsF1612.3 GB59details
RTX 4080 Super16 GBFitsF1612.3 GB61details
RTX 50608 GBFitsQ6_K5.7 GB90details
RTX 507012 GBFitsQ8_07.0 GB104details
RTX 508016 GBFitsF1612.3 GB79details
2x RTX 30902×24 GBFitsF1612.3 GB77details
2x RTX 40902×24 GBFitsF1612.3 GB83details
2x RTX 50902×32 GBFitsF1612.3 GB148details
RTX A600048 GBFitsF1612.3 GB63details
RTX 6000 Ada48 GBFitsF1612.3 GB79details
RTX PRO 6000 Blackwell96 GBFitsF1612.3 GB148details
L40S48 GBFitsF1612.3 GB71details
A100 40GB40 GBFitsF1612.3 GB128details
A100 80GB80 GBFitsF1612.3 GB168details
H100 PCIe80 GBFitsF1612.3 GB165details
H100 SXM80 GBFitsF1612.3 GB276details
H200141 GBFitsF1612.3 GB396details
RX 76008 GBFitsQ6_K5.7 GB58details
RX 7700 XT12 GBFitsQ8_07.0 GB67details
RX 7800 XT16 GBFitsF1612.3 GB51details
RX 7900 GRE16 GBFitsF1612.3 GB48details
RX 7900 XT20 GBFitsF1612.3 GB66details
RX 907016 GBFitsF1612.3 GB53details
RX 9070 XT16 GBFitsF1612.3 GB53details
Ryzen AI Max+ 395 128GB128 GBFitsF1612.3 GB21details
Arc A770 16GB16 GBFitsF1612.3 GB46details
Mac M1 16GB16 GBFitsQ8_07.0 GB11details
Mac M2 24GB24 GBFitsF1612.3 GB8.2details
Mac M3 24GB24 GBFitsF1612.3 GB8.2details
Mac M4 16GB16 GBFitsQ8_07.0 GB19details
Mac M4 32GB32 GBFitsF1612.3 GB9.9details
Mac M1 Pro 32GB32 GBFitsF1612.3 GB16details
Mac M3 Pro 36GB36 GBFitsF1612.3 GB12details
Mac M4 Pro 24GB24 GBFitsF1612.3 GB23details
Mac M4 Pro 64GB64 GBFitsF1612.3 GB23details
Mac M1 Max 64GB64 GBFitsF1612.3 GB33details
Mac M2 Max 96GB96 GBFitsF1612.3 GB33details
Mac M3 Max 128GB128 GBFitsF1612.3 GB33details
Mac M4 Max 64GB64 GBFitsF1612.3 GB45details
Mac M2 Ultra 192GB192 GBFitsF1612.3 GB66details
Mac M3 Ultra 256GB256 GBFitsF1612.3 GB68details

Questions

How much VRAM does Yi 6B need?

About 4.4 GB at 4-bit (Q4_K_M) with an 8K context: 3.4 GB of weights (real GGUF file size), 0.5 GB of KV cache and 0.5 GB of runtime overhead. At 8-bit it needs about 7.0 GB.

What is the smallest GPU that runs Yi 6B?

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 Yi 6B 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.

Model facts

Hugging Face repo01-ai/Yi-6B
Parameters6,061,035,520 (6.1B)
Layers / hidden size32 / 4096
AttentionGQA; 32 heads, 4 KV heads, head dim 128
KV cache per token (FP16)64.0 KiB
Max context4,096 tokens
Licenseapache-2.0
Released on HFNov 1, 2023
Downloads (30d) / likes16,514 / 376

Sources: Hugging Face model API · config.json (01-ai/Yi-6B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.