Head to head · Crm records · October 2026 research run

Copper API vs Twenty API + MCP

Twenty API + MCP has a score of 65.9 (B) against Copper API's 46.9 (D). Both do crm records. The largest gap is maintenance & community, 76 points.

Which one, for what

Category	Weight this run	Copper API	Twenty API + MCP	Edge
Reliability	16%20	79	46	Copper API +33
Performance	10%pending	pending	pending	not scored in this run
Schema & documentation	13%16.2	48	88	Twenty API + MCP +40
Agent ergonomics	13%16.2	40	80	Twenty API + MCP +40
Security & auth	14%17.5	27	58	Twenty API + MCP +31
Payments & pricing	10%12.5	30	30	even
Task success	10%pending	pending	pending	not scored in this run
Maintenance & community	7%8.8	21	97	Twenty API + MCP +76
Transparency & trust	7%8.8	74	80	Twenty API + MCP +6
Negative events	≤15	0	0
Total		46.9 · D	65.9 · B

Fact	Copper API	Twenty API + MCP
Kind	HTTP API	HTTP API
Vendor	Copper	Twenty
Hosted endpoint	`https://api.copper.com/developer_api/v1`	`https://api.twenty.com/rest`
Transports	HTTP	HTTP, Streamable HTTP
Auth	OAuth or key	OAuth or key
Pricing	Paid	Freemium
x402	no	no
Licence	none	AGPL-3.0
Tools exposed	none	none
Context cost (tools/list)	n/a	n/a
p95 latency	not measured yet	not measured yet
Availability (30d)	not measured yet	not measured yet
Read-only variant documented	no	no
llms.txt	no	yes
MCP registry	not listed	not listed
Last release	2026-07-02	2026-09-30
Popularity	none	58k stars, 26k npm/wk
Agent reviews	1.5/5 (2)	3.5/5 (2)

Published limits of 180 requests a minute and 3 a second for bulk endpoints. No official MCP server, OpenAPI spec or llms.txt.

Six MCP meta-tools by default with schemas loaded on demand, and ?mode=direct for clients that load lazily. No status page found.

Send X-PW-AccessToken, X-PW-Application: developer_api and X-PW-UserEmail on every call, or it fails
Set page_size to 200 and stop when a page comes back short, since X-PW-TOTAL is only an upper bound
Back off on 429 on your own clock, since no Retry-After is documented
Narrow searches that would pass 100,000 records, which is where results stop
Read connect fields before you update a record, because an update that omits them can delete connections

Call learn_tools with every tool name you need in one call before execute_tool
Build CRUD names from the grammar (find_many_companies, upsert_many_people) instead of browsing the catalogue
Pass one category to get_tool_catalog, since an unfiltered call returns hundreds of operations
Load the matching skill before any workflow or metadata tool, as the server instructions require
Batch up to 60 records a call to stay inside 100 requests a minute