One scale for every product
Pebbles Ai is scored on the same axes and the same 0 to 100 scale as everyone else.
AI assistants compared: the GTM landscape
How 25 AI assistants compare on models, tools, memory and controlled actions. Editorial scores from Pebbles Ai's research, with Pebbles Ai on the same scale.
AI assistants
Models, tools, memory and controlled actionsPebbles Ai on each axis
- Business context80 out of 100Highest rival 90
- Knowledge-work breadth70 out of 100Highest rival 90
- Business-action breadth70 out of 100Highest rival 90
- Work controls70 out of 100Highest rival 80
- Connected GTM workflow80 out of 100Highest rival 100
- User-set autonomy80 out of 100Highest rival 90
Positioning map
Pebbles Ai
Business context across, Business-action breadth up, both 0 to 100. Arrow keys move between products.
Not scored on these axes: Hermes Agent
Every axis
Pebbles AiOther productsMedian
Business context
What it measures
Built-in company knowledge and GTM objects available to the assistant.
80Not scored: Hermes Agent
Knowledge-work breadth
What it measures
Range of non-coding research, analysis, writing and artifact work.
70Not scored: Hermes Agent
Business-action breadth
What it measures
Native ability to perform actions across business systems; coding actions excluded.
70Not scored: Hermes Agent
Work controls
What it measures
Planning, approvals, persistence and recovery.
70Not scored: Hermes Agent, Manus
Connected GTM workflow
What it measures
How many of eight go-to-market stages run natively in one workspace: account research, lead discovery, CRM records, grounded drafting, mailbox send, social publishing, calendar actions and memory.
80User-set autonomy
What it measures
Whether the user chooses how far the AI acts on its own, from asking before each action to acting, and how widely that choice applies across mail, CRM, calendar, social and connected tools.
80
Score matrix
0 to 100, Pebbles Ai pinned first
| Pebbles Ai | Pebbles Ai | 80 | 70 | 70 | 70 | 80 | 80 |
|---|---|---|---|---|---|---|---|
| ChatGPT work and agent experiences | Suite | 60 | 90 | 80 | 70 | 60 | 90 |
| Claude and Artifacts | Suite | 60 | 90 | 50 | 60 | 60 | 90 |
| Claude Code | Adjacent | 20 | 30 | 20 | 80 | 30 | 90 |
| Dust | Direct | 80 | 70 | 70 | 70 | 60 | 80 |
| OpenAI Codex | Adjacent | 20 | 30 | 20 | 80 | 10 | 80 |
| Glean | Direct | 90 | 70 | 50 | 70 | 60 | 70 |
| Lindy | Direct | 60 | 60 | 80 | 50 | 70 | 70 |
| Microsoft Copilot Studio | Suite | 70 | 60 | 90 | 80 | 30 | 70 |
| Relevance AI | Direct | 70 | 70 | 80 | 60 | 70 | 70 |
| Salesforce Agentforce | Suite | 90 | 60 | 90 | 80 | 80 | 70 |
| HubSpot Breeze | Suite | 90 | 60 | 80 | 70 | 100 | 60 |
| Zapier Agents | Direct | 50 | 50 | 90 | 50 | 80 | 60 |
| CrewAI | Open source or DIY | 40 | 60 | 50 | 50 | 10 | 50 |
| Hermes Agent | Open source or DIY | Not scored | Not scored | Not scored | Not scored | 40 | 50 |
| Microsoft 365 Copilot | Suite | 90 | 80 | 90 | 80 | 60 | 50 |
| Google Gemini for Workspace | Suite | 80 | 80 | 70 | 70 | 60 | 40 |
| Perplexity | Direct | 50 | 80 | 30 | 30 | 50 | 40 |
| Writer | Direct | 90 | 80 | 70 | 80 | 60 | 40 |
| Manus | Direct | 50 | 80 | 60 | Not scored | 50 | 30 |
| OpenWebUI plus tools | Open source or DIY | 30 | 50 | 30 | 30 | 30 | 20 |
| Pi coding agent | Adjacent | 20 | 20 | 10 | 40 | 10 | 20 |
| Dify | Open source or DIY | 40 | 60 | 50 | 50 | 0 | 10 |
| LangGraph | Open source or DIY | 30 | 50 | 50 | 80 | 0 | 10 |
| Microsoft AutoGen | Open source or DIY | 30 | 50 | 40 | 40 | 0 | 10 |
| n8n AI | Open source or DIY | 40 | 50 | 80 | 60 | 0 | 10 |
Swipe to see every axis
Scores are editorial estimates (0 to 100) from Pebbles Ai's own research, not measured benchmarks.
How to read the landscape
Compare within a direction
Each subdirection has four axes of its own, plus two shared by all. Read a score against its neighbours only.
A blank is not a zero
Where a score is missing, the research did not rate that product on that axis.
Share or cite this page
Copy the link or a ready-made citation, so your readers can find the original.
AI assistants compared: the GTM landscape Pebbles Ai https://trypebbles.ai/resources/gtm-competitive-landscape/assistants
Questions about AI assistants
What does the AI assistants landscape compare?
It compares 25 AI assistants on models, tools, memory and controlled actions. Every product sits on the same map, and Pebbles Ai is scored beside them on the same scale.
Which axes are AI assistants scored on?
Each product is scored on six axes. The first four are Business context, Knowledge-work breadth, Business-action breadth and Work controls. The last two, Connected GTM workflow and User-set autonomy, are shared by every direction, so you can compare the same things across the whole landscape.
Where does Pebbles Ai stand in AI assistants?
Pebbles Ai has no top score in AI assistants. Its strongest axis is User-set autonomy, where it scores 80 against 90 for the best product. The map shows which products lead on each axis.
How are the AI assistants scores produced?
They are editorial estimates from Pebbles Ai's own research, on a scale from 0 to 100. They are not measured benchmarks or rankings. A blank cell means a product was not scored on that axis, not that it scored zero. Pebbles Ai is scored on the same scale as every other product.
Your next chapter
See where Pebbles Ai fits your stack
Bring the job you need done. Compare it against the directions that matter to your team.
