Dolphin GGUF models

95 GGUF models in the Dolphin family, quantized by 34 uploaders. Sizes run from 386M to 69B parameters. The longest trained context is 1024K tokens.

TinySmallMediumLarge
Most downloadedNewestLargestLongest contextSmallest
DO
Dolphin3 Cyber 8B
Dolphin · RavichandranJ
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
DO
Dolphin X1 8B
Dolphin · dphn
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
DO
weyaxi_dolphin2.1-openorca-7b
Dolphin · TheBloke
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
DO
Dolphin3 Cyber 8B
Dolphin · meowmeow123ok
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
DO
Dolphin3 Cyber 8B
Dolphin · ScottzillaSystems
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
DO
Dolphin v2
Dolphin · prithivMLmods
Params3.1B
Context125K
Q4_K_M1.9 GB
CodingVision
DO
Dolphin v2 8b Abliterated
Dolphin · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
GGUF
DO
Dolphin 2.1 70B
Dolphin · TheBloke
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
DO
Dolphin 2.2 70b
Dolphin · mradermacher
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
DO
Dolphin Email 8b
Dolphin · mradermacher
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
DO
dolphin 2.2 70B
Dolphin · TheBloke
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
DO
Dolphin 2.6 FC Hermes 2 Pro
Dolphin · mradermacher
Params7.24B
Context32K
Q4_K_M4.5 GB
Tool use
DO
Dolphin 2.6 FC Hermes 2 DPO
Dolphin · mradermacher
Params7.24B
Context32K
Q4_K_M4.5 GB
Tool use
DO
Dolphin3 Cyber 8B
Dolphin · V12nm
Params8.03B
Context128K
Q4_K_M4.9 GB
GGUF
DO
dolphin-ultrafeedback-dpo
Dolphin · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
DO
Dolphin v2
Dolphin · mradermacher
Params3.09B
Context125K
Q4_K_M1.9 GB
CodingTool useVision
DO
Dolphin 2.2 70b
Dolphin · mradermacher
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
DO
Dolphin3 Cyber 8B
Dolphin · nbpedro315
Params8.03B
Context128K
Q4_K_M4.9 GB
GGUF
DO
Dolphin X1 8B
Dolphin · mradermacher
Params8.03B
Context128K
Q4_K_M4.9 GB
Tool use
DO
dolphin-cn-dialect-small-streaming
Dolphin · cstr
Params386M
Context—
Smallest0.3 GB
GGUF
DO
Dolphin Email 8b
Dolphin · mradermacher
Params8.03B
Context128K
Q4_K_M4.9 GB
Tool use
DO
Dolphin 2.1 70b
Dolphin · mradermacher
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
DO
Dolphin X1 Trinity Nano
Dolphin · mradermacher
Params6.12B
Context128K
Q4_K_M3.8 GB
Tool use
DO
Dolphin v2
Dolphin · mradermacher
Params3.09B
Context125K
Q4_K_M1.9 GB
CodingTool useVision
DO
Dolphin 2.7 Yayi2 30b
Dolphin · mradermacher
Params30.4B
Context8K
Q4_K_M18.2 GB
GGUF
DO
Dolphin 2.8 Experiment26 7b
Dolphin · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
CodingTool use
DO
Dolphin3 Cyber 8B
Dolphin · kepom
Params8.03B
Context128K
Q4_K_M4.9 GB
GGUF
DO
Dolphin Xgen RL
Dolphin · mradermacher
Params10.7B
Context256K
Q4_K_M6.5 GB
GGUF
DO
dolphin-ultrafeedback-dpo
Dolphin · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
DO
dolphin 2.8 experiment26 7b
Dolphin · bartowski
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF

About the Dolphin GGUF models

Dolphin is represented on ggufmodels.com by 95 GGUF models, quantized by 34 uploaders on Hugging Face. The most downloaded are Dolphin3 Cyber 8B, Dolphin X1 8B and weyaxi_dolphin2.1-openorca-7b. Parameter counts run from 386M to 69B, and the longest trained context window is 1024K tokens.

A Q4_K_M file of a Dolphin model is between 1.9 to 41.4 GB, which is the quant most people start with. Most files come from mradermacher (50), harkov000 (4) and RichardErkhov (4). The quants you will meet most often are Q4_K_M (79), Q5_K_M (77), Q2_K (70) and Q6_K (69), where the number is how many models offer that quant.

Frequently asked questions

Which Dolphin GGUF should I download?

The most downloaded Dolphin GGUF here is Dolphin3 Cyber 8B. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.

How much VRAM does a Dolphin model need?

Q4_K_M files of Dolphin models on this site range from 1.9 to 41.4 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.

Where do Dolphin GGUF files come from?

The most common quantizers here are mradermacher, harkov000 and RichardErkhov. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.

Keep browsing