Pass source google images for visual demand, or source google news for news-search demand. Same weekly grain, no volume on either, and 7D signs that flip on chatgpt, nike, and bitcoin.
Live data as of 2026-10-21
A google images versus google news comparison on Trends MCP is two keyword sources on get_growth and get_time_series. Source google images returns a weekly 0-100 visual-demand series for any phrase. Source google news returns a weekly 0-100 Google News search series for the same phrase. Neither close printed a volume field. Both tesla series on October 21, 2026 held 261 weekly points from 2021-10-23 through 2026-10-17. google images closed 53.0. google news closed 100.0. chatgpt, nike, and bitcoin flipped the 7D sign: images fell, news rose. There is no Images live board. type google news is a 400. The live boards use type Google Trends and type Google News Top News.
The October 21, 2026 pull is the split that still matters next month: same weekly grain, same close date, then a 53.0 versus 100.0 close on tesla. get_time_series for tesla on source google images returned 261 weekly points from 2021-10-23 (value 38) through 2026-10-17 (value 53). Peak 100 sat on 2026-04-04. No row carried a volume key. Growth metadata printed total_data_points 261.
get_time_series for tesla on source google news returned 261 weekly points from 2021-10-23 (value 6) through 2026-10-17 (value 100). Peak 100 sat on that last week. An earlier news spike sat at 66 on 2026-08-22. No row carried a volume key. Peak week is not shared. Images peaked in April 2026. News peaked on the close week.
Growth windows line up by preset name and split by size. google images tesla 7D matched 1W: 53.0 from 30.0, plus 76.67 percent, recent_date 2026-10-17, baseline 2026-10-10. 30D matched 1M: 53.0 from 40.0, plus 32.5 percent. 12M was plus 130.43 percent from 23.0. YTD was the same plus 130.43 percent from 23.0 on 2026-01-03. google news tesla 7D matched 1W: 100.0 from 32.0, plus 212.5 percent. 30D matched 1M: 100.0 from 10.0, plus 900.0 percent. 12M was plus 2400.0 percent from 4.0. YTD was the same plus 2400.0 percent from 4.0. Omitting percent_growth defaults to 12M on both and succeeded on both. REST source Google Images case-folded and echoed google images with the same 12M row.
| Keyword | Images close | Images 7D | Images 12M | News close | News 7D | News 12M | Points |
|---|---|---|---|---|---|---|---|
| tesla | 53.0 on 2026-10-17 | +76.67% from 30.0 | +130.43% from 23.0 | 100.0 on 2026-10-17 | +212.5% from 32.0 | +2400.0% from 4.0 | 261 / 261 |
| chatgpt | 80.0 on 2026-10-17 | -4.76% from 84.0 | -12.09% from 91.0 | 70.0 on 2026-10-17 | +89.19% from 37.0 | +600.0% from 10.0 | 261 / 261 |
| nike | 53.0 on 2026-10-17 | -7.02% from 57.0 | +1.92% from 52.0 | 66.0 on 2026-10-17 | +78.38% from 37.0 | +1550.0% from 4.0 | 261 / 261 |
| bitcoin | 6.0 on 2026-10-17 | -25.0% from 8.0 | -53.85% from 13.0 | 91.0 on 2026-10-17 | +97.83% from 46.0 | +2933.33% from 3.0 | 261 / 261 |
| spiderman | 39.0 on 2026-10-17 | +2.63% from 38.0 | +85.71% from 21.0 | 3.0 on 2026-10-17 | -40.0% from 5.0 | -40.0% from 5.0 | 261 / 261 |
chatgpt Images first non-zero week was 2022-12-10 at value 1. Peak 100 sat on 2025-09-20. Keyword ChatGPT echoed search_term ChatGPT and matched the lowercase 12M row on both sources: Images 80.0 from 91.0, minus 12.09 percent; News 70.0 from 10.0, plus 600.0 percent. Casing folded. It did not create a second series.
spiderman is the inverse of tesla. Images closed 39.0. News closed 3.0. Visual demand stayed after the news-search index went quiet. bitcoin is the other extreme: Images 6.0 with 12M minus 53.85 percent, News 91.0 with 12M plus 2933.33 percent from 3.0. Quote both closes. Do not rescale one source onto the other.
A live Google Trends title is a spike, not a five-year comparison. Rank 1 on type Google Trends was you can see everything. Rank 2 was tesla roadster. That string closed 100.0 on both sources for 7D. Images 7D was 100.0 from 5.0, plus 1900.0 percent. News 7D was 100.0 from 4.0, plus 2400.0 percent. News 12M printed 999999.0 from baseline 0.0, with data_quality low (over 90 percent zero values). Quote the 0-100 close. Do not chart 999999.0 as a percent of News coverage.
Source names case-fold. Feed labels do not belong here. Source Google Trends returned HTTP 400 invalid_source, message 'google trends' is not a valid source., and listed 16 keyword sources plus all. The handler lowercased the rejected type string. It did not map it to google images or google news.
google images tesla printed no volume. google news tesla printed none. Those are two empty volume payloads on the same keyword, which is the opposite of a google search mix that can weight estimated volume.
{"mode": "get_growth", "source": "google images, google news", "keyword": "tesla", "percent_growth": ["12M"]}
That comma call on October 21, 2026 returned sources_successful 2. trend_analysis printed uptrend_count 2, downtrend_count 0, and uptrend_ratio 2/2. Both 12M windows rose. volume_weighted_growth was null, has_volume_data false, sources_with_volume 0, weight_coverage 0/2 sources, and total_volume 0. The mixer had nothing to weight. Read source_results for the 53.0 versus 100.0 closes. Do not treat uptrend_ratio 2/2 as proof the indexes moved by the same amount.
A first six-window comma call on tesla returned sources_successful 1. google news printed unavailable with message Data not available for this source. Dedicated get_growth on source google news with the same six windows succeeded, and the later 12M comma succeeded both. Retry a failed source on its own call before treating a comma miss as empty history.
Spaces in the comma string are optional. Source google images,google news on tesla 12M matched the spaced call: sources_successful 2, uptrend_ratio 2/2, volume_weighted_growth null. REST google images,google news on nike 7D also returned sources_successful 2. get_time_series rejects the comma either way. Source google images, google news returned HTTP 400 invalid_source and listed 16 keyword sources with no all. One source per series call. Two sources belong on get-growth.
Two strings, two tools. REST uses mode get_growth or get_top_trends. MCP uses the matching tool names. Image windows live on the Google Images trends API. News-search history lives on the Google News keyword trends API. Headline follow-up lives on source google news versus type Google News Top News.
{"mode": "get_growth", "source": "google images, google news", "keyword": "tesla", "percent_growth": ["7D", "30D", "12M"]}
{"mode": "get_time_series", "source": "google images", "keyword": "tesla"}
{"mode": "get_time_series", "source": "google news", "keyword": "tesla"}
{"mode": "get_top_trends", "type": "Google Trends"}
{"mode": "get_top_trends", "type": "Google News Top News"}
The MCP equivalents are get_growth(keyword="tesla", source="google images, google news", percent_growth=["7D", "30D", "12M"]), get_time_series(keyword="tesla", source="google images"), get_time_series(keyword="tesla", source="google news"), get_top_trends(type="Google Trends"), and get_top_trends(type="Google News Top News"). REST 1W matched 7D on tesla for both sources. REST 1M matched 30D on both.
Omitting limit on type Google Trends returned limit 25, count 25, offset 0, and as_of_ts 2026-10-21T06:01:38.601407+00:00. Rank 1 was you can see everything. Rank 2 was tesla roadster. Rank 3 was oregon shooting. Those rows are search phrases. Omitting limit on type Google News Top News is a different board. Rank 1 was an NPR California list headline. Rank is live story order, not a sort of the 2026-10-17 News index.
REST type google search (the keyword source string, lowercase) returned HTTP 200, echoed type Google Trends, and matched you can see everything at rank 1 and tesla roadster at rank 2. REST type google news top news echoed type Google News Top News and matched the NPR rank 1 headline. REST type google news (the keyword source string) returned HTTP 200 with envelope statusCode 400, error invalid_request, message Unknown type 'google news', plus 24 valid feed labels. REST type google images failed the same way. llms.txt still describes feed labels as case-sensitive. The live handler accepted some lowercase aliases on October 21, 2026 and rejected others. Prefer the canonical spelling in stored configs. Do not pass source google news or source google images as a feed type.
REST omitting type is a different 400. POST https://api.trendsmcp.ai/api with mode get_top_trends and limit 5, no type, returned HTTP 200 with envelope statusCode 400, error invalid_request, message The 'type' parameter is required, then the same 24-name list. llms.txt still names that miss missing_parameter. Read statusCode inside the JSON, not only the HTTP line.
get_top_trends MCP pricing is 1 credit per 10 requested rows, max 10 credits per call. REST counts 1 request per call. Free boards still cap at the top 10. The 25-row Google Trends pull on this page is the paid default, which is also what the handler returns when limit is omitted on a paid key. Keyword follow-up stays on source google images or source google news. The live boards stay on their type strings.
FAQ