Skip to content
ADE Index

Crush vs Hermes Agent vs OpenCode

Up to four tools, row by row. We do not guess: a cell we have not checked says so.

How they differ

Pick this if…

Paraphrased from each maker’s own description. Select a line for the quote.

Checked for Hermes Agent and OpenCode, not yet for Crush: Price (Hermes Agent: ; OpenCode: )

Checked for Crush and OpenCode, not yet for Hermes Agent: Systems (Crush: , ; OpenCode: , )

Not yet checked for any of these tools: agents it runs.

Dotted values open their quote.

What each can do

Select an answer to see the maker’s sentence, or what we searched.

We have not checked Crush’s capabilities yet. 24 capabilities are checked for at least two of these tools. 5 differ only in a limit. 8 others are checked for one tool only.

Differences only5
  • Yesthe maker documents it
  • *its limit or add-on is named under the row
  • Emptynot yet checked
  • From the listingthe listing’s own facts; not one of the capabilities we check

Models, keys and usage

Bring your own keyFrom the listing
Named model providersFrom the listing
,,,,,,,,
+26 more model providers
,,,,,,,,,,,,,,,,,,,,,,,,,
and others; list may be partial
Listing also names 8 othersSpaceXSI (Grok), LM Studio, Moonshot AI, Alibaba (Qwen), OpenCode Console (OpenCode Zen), Nebius Token Factory (Nebius AI Studio), Neuralwatt Cloud, Qiniu Cloud
,,,,,,,,
+30 more model providers
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Listing also names 11 othersAmazon Bedrock, SpaceXSI (Grok), Moonshot AI, Vercel AI Gateway, Alibaba (Qwen), OpenCode Console (OpenCode Zen), Microsoft Foundry (Azure AI Foundry), Meta AI (Llama), Nebius Token Factory (Nebius AI Studio), Novita AI, Command Code
,,,,,,,,
+2 more model providers
,
and others; list may be partial
Listing also names: SpaceXSI (Grok), GitHub Copilot
Local modelsNot yet checked: Crush
Not yet checked for Crush
Custom OpenAI-compatible endpointNot yet checked: Crush
Not yet checked for Crush
Token and cost usageNot yet checked: Crush
Not yet checked for Crush

MCP, hooks, skills and subagents

MCP serversNot yet checked: Crush
Not yet checked for Crush
local and remote
local and remote
HooksNot yet checked: Crush
Not yet checked for Crush
SkillsNot yet checked: Crush
Not yet checked for Crush
SubagentsNot yet checked: Crush
Not yet checked for Crush
PluginsNot yet checked: Crush
Not yet checked for Crush
Custom commandsNot yet checked: Crush
Not yet checked for Crush

Editors, scripts and CI

Editor integrationNot yet checked: Crush
Not yet checked for Crush
ACP editors
ACP clients such as Zed
Headless modeNot yet checked: Crush
Not yet checked for Crush
JSON outputNot yet checked: Crush
Not yet checked for Crush

Review, steer and undo

Plan modeNot yet checked: Crush, Hermes Agent
Not yet checked for Crush
Not yet checked for Hermes Agent
Read-only modeNot yet checked: Crush, Hermes Agent
Not yet checked for Crush
Not yet checked for Hermes Agent
Plan profile; shell still asks
Review diffsNot yet checked: Crush
Not yet checked for Crush
Needs checkpoints on
In ACP permission requests
Undo file editsNot yet checked: Crush
Not yet checked for Crush
Opt-in; local and SSH only

Parallel agents

Parallel sessionsNot yet checked: Crush
Not yet checked for Crush
Git worktreesNot yet checked: Crush
Not yet checked for Crush

Context and memory

Project instruction filesNot yet checked: Crush
Not yet checked for Crush
Memory across sessionsNot yet checked: Crush, OpenCode
Not yet checked for Crush
Not yet checked for OpenCode
Resume a sessionNot yet checked: Crush
Not yet checked for Crush
Browse past sessionsNot yet checked: Crush
Not yet checked for Crush

Remote and mobile control

Remote controlNot yet checked: Crush, OpenCode
Not yet checked for Crush
Not yet checked for OpenCode
Steer from a phoneNot yet checked: Crush, OpenCode
Not yet checked for Crush
Unblocks Kanban workers
Not yet checked for OpenCode
NotificationsNot yet checked: Crush
Not yet checked for Crush

Sandbox, permissions and privacy

Approval modesNot yet checked: Crush
Not yet checked for Crush
Runs in a containerNot yet checked: Crush
Not yet checked for Crush
Official Docker setup
Official Docker image

Images, web and your computer

Image inputNot yet checked: Crush
Not yet checked for Crush
Needs an image-capable model
Fetch URLsNot yet checked: Crush, Hermes Agent
Not yet checked for Crush
Not yet checked for Hermes Agent
Browser automationNot yet checked: Crush, OpenCode
Not yet checked for Crush
Not yet checked for OpenCode
Voice inputNot yet checked: Crush, OpenCode
Not yet checked for Crush
Not yet checked for OpenCode
Not yet checked: , ,

5 capabilities are not yet checked for any of these tools and are not shown. Each answer is about the CLI agent in its column. Something the maker publishes separately, such as an editor plugin or a GitHub Action, shows as “With add-on”, even when it has its own listing here. How capabilities are checked

About these tools

Hermes Agent

OpenCode

How to read the Index score

At a glance

Index score out of 100

Crush23Hermes Agent48OpenCode79

Mindshare HN mentions, 90 days

Crush7Hermes Agent84OpenCode1,414

Adoption GitHub stars, not scored

Crush29.1kHermes Agent264kOpenCode228k

Release days last 90 days

Crush16Hermes Agent19OpenCode28

Lowest monthly plan billed monthly, lower is cheaper

CrushNot yet checkedHermes AgentNot yet checkedOpenCode$10/month

Platforms systems and surfaces

Crush2Hermes AgentNot yet checkedOpenCode3

Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only. After activation, the score is 80% mindshare plus 20% of the greater of mindshare and the vote score. GitHub stars do not affect the Index score. Mindshare uses a log scale; votes use the square-root curve and a 100-vote floor. Ranks use unrounded scores. Mindshare and adoption bars are log-scaled; the others are linear within this comparison. How the score is built

Differences only
3 of 4
Crushcharmbracelet · CLI agent
Hermes AgentCLI agent
OpenCodeAnomaly · CLI agent
Interface
Crush TUI viewing a Go file while thinking, with session sidebar showing model, cost and LSPs
Terminal UI with an inline diff, passing bun tests and a context sidebar
What it is
Not yet checked
Terminal agent for writing, reviewing and running code in local projects
Coding agent with switchable build and read-only plan agents
Category
CLI agent
CLI agent
CLI agent
Pricing
Starts at (billing shown)
Not yet checked
Lowest monthly plan
Not yet checked
Not yet checked
Free tier
Not yet checked
Plans
Not yet checked
Runs on
Android
Yes
Not yet checked
Not yet checked
Terminal
Yes
Not yet checked
Yes
Web
Not yet checked
Not yet checked
Yes
Windows
Not yet checked
Not yet checked
Yes
Models and agents
Drives these agents
Not yet checked
Not yet checked
Not yet checked
Source and activity
Source
GitHub stars
Most
Release days, last 90 days
16
19
28Most
Release days per week
Latest release, any channel
nightly 18 Nov 2026
v2.3.2 18 Nov 2026
Index score
Score and rank
23 #64 Score 22.6; ranked on the unrounded score
48 #14 Score 48.2; ranked on the unrounded score
79 #3 Score 78.8; ranked on the unrounded score
Mindshare
237 Hacker News mentions in the last 90 days
4884 Hacker News mentions in the last 90 days
791,414 Hacker News mentions in the last 90 days
Adoption, not scored
8429,112 stars, read 18 Nov 2026
100264,230 stars, read 18 Nov 2026
100227,631 stars, read 18 Nov 2026
Votes
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
Rank history
First ranked 18 Nov 2026
First ranked 18 Nov 2026
First ranked 18 Nov 2026
Full page

Watch Crush vs Hermes Agent vs OpenCode on YouTube

More comparisons

Build your own