Ant Group/Mixture of Experts

ALing 2.6 Flash

Name: Ling 2.6 Flash
Author: Ant Group

Today, we announce the official open-source release of Ling-2.6-flash, an instruct model with 104B total parameters and 7.4B active parameters.

chatcodingreasoningmultilingual

104B

Parameters (7.4B active)

128K

Context length

Benchmarks

Quantizations

Architecture

MoE

Released

2026-02-10

Layers

KV Heads

Head Dim

128

Family

other

Quantization Options

Quant	Bits	VRAM	Quality
IQ2_XXS	2.38	31.4 GB	low
IQ2_M	2.93	38.6 GB	low
Q2_K	3.16	41.6 GB	low
IQ3_XXS	3.25	42.7 GB	low
IQ3_XS	3.5	46.0 GB	low
Q3_K_S	3.64	47.8 GB	low
IQ3_M	3.76	49.4 GB	low
Q3_K_M	4	52.5 GB	low
Q3_K_L	4.3	56.4 GB	moderate
IQ4_XS	4.46	58.5 GB	moderate
Q4_K_S	4.67	61.2 GB	moderate
Q4_K_M	4.89	64.1 GB	good
Q5_K_S	5.57	72.9 GB	good
Q5_K_M	5.7	74.6 GB	good
Q6_K	6.56	85.8 GB	excellent
Q8_0	8.5	111.0 GB	lossless
FP16	16	208.5 GB	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN LING 2.6 FLASH NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (9)

τ²-Bench86.0

GPQA Diamond59.3

IFBench57.4

SciCode27.1

AA Intelligence26.2

AA Long Context25.0

AA Coding23.2

Terminal-Bench21.2

HLE6.2

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run other:104b-q4_K_M

Tag may need adjustment — check ollama.com/library/other for available tags.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace GGUF Downloads Build Hardware