Skip to main content

Edge0 AILanguage modelWeights available

Edge0-120B

PopularityRank 193 of 332 language models

36out of 100first value

usage data only, range 23 to 50

Summary

Edge0-120B is a language model for local coding and agent tasks. Edge0 reports that it uses 8.64 billion of its 120 billion parameters per token.

The model card lists 27.98 GiB for the Q3_K weights and 32.96 GiB for NVFP4. Running the model also needs memory for inputs and computation.

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

The 100 most popular models of this category, 0 to 100, higher is more popular

20 of 101 models

As of November 2026 · Source: gradually.ai popularity index

Over time

First value from November 15, 2026. The timeline appears from the second week.

What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.

All specifications

Architecture

Architecture
48 layers with 512 experts each.
Source: Architecture

Input and output

Input
Text
Source: Input
Output
Text
Source: Output
Context window
262,144 tokens
Source: Context window
Context limit in the launch template
The launch template caps context at 131,072 tokens. The model configuration lists 262,144 tokens.
Source: Context limit in the launch template

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Weights license
Edge0 Community Model License 1.0. Qwen components remain subject to the Qwen Community License 1.0.
Source: Weights license
Shutdown
None announced
no source

Further details

Release announcement
Edge0 announced the release on November 12, 2026.
Source: Release announcement
Release log
The Edge0 repository lists the release under November 13, 2026.
Source: Release log
Q3_K weights
Edge0/Edge0-120B-q3k
Source: Q3_K weights
NVFP4 weights
Edge0/Edge0-120B-NVFP4
Source: NVFP4 weights
Tested hardware
Edge0 ran the Q3_K weights locally on an RTX 5090 D with a test budget of 16 GiB of video memory and 20 GiB of system RAM. These figures describe the test setup.
Source: Tested hardware
Commercial deployment
Selling the model, running a commercial model service, and offering standalone commercial coding or office assistants require a separate written license from Edge0.
Source: Commercial deployment

Frequently asked questions about Edge0-120B

How large is the context window of Edge0-120B?

262,144 tokens. That is how much text Edge0-120B reads at once, about 390 A4 pages.

Page count as a rule of thumb from Artificial Analysis

Context windowTokens the model reads at once · higher is better
Edge0-120BRank 77 of 112262K
Kimi K2.6262K
Qwen 3.8 27B262K
Kimi K2.7 Code262K
Kimi K2.5262K
Kimi K2 Thinking262K
Mistral Medium 3.5262K

Can I run Edge0-120B locally?

Yes, the weights are open. Edge0 ran the Q3_K weights locally on an RTX 5090 D with a test budget of 16 GiB of video memory and 20 GiB of system RAM. These figures describe the test setup.

Stated by Edge0 AI

Open weights
yes
License
Edge0 Community Model License 1.0. Qwen components remain subject to the Qwen Community License 1.0.
Parameters
120B
active per token
8.6B