Mira GGUF models

123 GGUF models in the Mira family, quantized by 13 uploaders. Sizes run from 3M to 69B parameters. The longest trained context is 256K tokens.

TinySmallMediumLarge
Most downloadedNewestLargestLongest contextSmallest
MI
Mira 12b
Mira · mradermacher
Params11.9B
Context256K
Q4_K_M7.4 GB
ReasoningRoleplayTool use
MI
mira agent instinct
Mira · Hoodx
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
MI
Mira 1 Large 0919
Mira · mradermacher
Params14.8B
Context40K
Q4_K_M9.0 GB
CodingReasoningTool use
MI
Mira v1.23 27B Rlvr
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira Agent Instinct
Mira · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
MI
Mira v1.12.1 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Coding
MI
Mira v1.19 27B Dpo
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira v1.3 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira 12b
Mira · mradermacher
Params11.9B
Context256K
Q4_K_M7.4 GB
ReasoningRoleplayTool use
MI
Mira 1.9 Alternate 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.7 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.9 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira M2
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira 1.10 Dpo 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.22.1 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.21 27B Rlvr
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira V1.8.1B 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
CodingVision
MI
Mira 1.9 Alternate 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira v1.22.1 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira 1 Large
Mira · mradermacher
Params14.8B
Context40K
Q4_K_M9.0 GB
CodingReasoningTool use
MI
Mira v1.2 Dpo 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Coding
MI
Mira V1.22.2B 27B Dpo
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira Demeter
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.19 27B Dpo
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.20 27B Dpo
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
MI
Mira v1.25 27B Wave
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira v1.12.1 27B
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
CodingVision
MI
Mira v1.21 27B Rlvr
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision
MI
Mira v1.25.1 27B DPO
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Tool use
MI
Mira Demeter
Mira · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
Vision

About the Mira GGUF models

Mira is represented on ggufmodels.com by 123 GGUF models, quantized by 13 uploaders on Hugging Face. The most downloaded are Mira 12b, mira agent instinct and Mira 1 Large 0919. Parameter counts run from 3M to 69B, and the longest trained context window is 256K tokens.

A Q4_K_M file of a Mira model is between 0.8 to 41.4 GB, which is the quant most people start with. Most files come from mradermacher (101), virtuous7373 (6) and gotzmann (4). The quants you will meet most often are Q4_K_M (105), Q5_K_M (102), Q5_K_S (100) and Q4_K_S (100), where the number is how many models offer that quant.

Frequently asked questions

Which Mira GGUF should I download?

The most downloaded Mira GGUF here is Mira 12b. 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 Mira model need?

Q4_K_M files of Mira models on this site range from 0.8 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 Mira GGUF files come from?

The most common quantizers here are mradermacher, virtuous7373 and gotzmann. 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