DGrid AI Heat Score: history and how it is built

The Heat Score measures how much real human conversation a coin has compared with its own recent past. It is not price, not market cap and not raw engagement.

Heat Score now

62/100

Hot

Contributors 24h

vs 7d median 22 (2.5×)

Contributors 30d median

7d vs 30d 0.6×

Interactions 24h

11.1K

vs 7d median 10.9K (1.0×)

Spam share

7.5%

penalty above 15%

How Heat Score v0 works

Start at 50. Add points when today's unique contributors beat their 7-day median (weight 22 per doubling) and when the last 7 days beat the 30-day median (weight 12); do the same for interactions with lower weights (8 + 8). Subtract a penalty when the share of posts flagged as spam exceeds 15%. Clamp between 0 and 100. Labels: Cold <35, Warm 35-54, Hot 55-74, On fire 75+.

Why contributors first? A handful of accounts can farm millions of interactions; thousands of distinct people posting is much harder to fake. Contributor and spam data come from LunarCrush.

Heat Score — 48 days

Daily contributors

Daily interactions

Daily sentiment (0-100)

Last 14 days

Date	Contributors	Interactions	Posts	Spam	Sentiment	Heat
2026-10-16	53	10.9K	68	8.8%	100	61
2026-10-15	53	32.6K	59	5.1%	100	74
2026-10-14	37	22.5K	43	7.0%	100	60
2026-10-13	22	2.6K	30	10.0%	100	20
2026-10-12	18	31.0K	21	14.3%	100	40
2026-10-11	17	4.5K	18	22.2%	100	11
2026-10-10	13	1.7K	13	46.2%	100	0
2026-10-09	40	831	42	16.7%	97	21
2026-10-08	34	11.6K	38	18.4%	100	46
2026-10-07	59	4.2K	66	9.1%	98	59
2026-10-06	21	9.7K	24	29.2%	83	30
2026-10-05	31	21.6K	34	20.6%	80	57
2026-10-04	41	6.6K	44	20.5%	63	51
2026-10-03	36	24.5K	38	26.3%	65	59