Model Freshness Tracker

Analyzes how recently AI models and their data have been updated. Track freshness scores, identify stale data, and see which providers keep their models most up to date.

Avg Freshness

100.0

out of 100

Last Hour

293

models updated

Last 24h

293

models updated

Stalest Model

GPT-5.4 Pro

0m ago

Freshness Tier Distribution

Models grouped by how recently their data was last updated.

Just Updated

293100.0%

Fresh

00.0%

Recent

00.0%

Aging

00.0%

Stale

00.0%

Freshest Models

Top 20

Models with the most recently updated data, ranked by freshness score.

#	Model	Provider	Rank	Last Updated	Freshness
1	Inflection 3 Pi	Inflection	#280	0m ago	100
2	Inflection 3 Productivity	Inflection	#281	0m ago	100
3	Qwen2.5-VL 7B Instruct	Alibaba	#282	0m ago	100
4	Mistral Large	Mistral AI	#283	0m ago	100
5	GPT-3.5 Turbo (older v0613)	OpenAI	#284	0m ago	100
6	Gemma 2 27B	Google	#285	0m ago	100
7	Llama 3 70B Instruct	Meta	#286	0m ago	100
8	Llama 3.2 3B Instruct	Meta	#287	0m ago	100
9	WizardLM-2 8x22B	Microsoft	#288	0m ago	100
10	Llama 3.2 1B Instruct	Meta	#289	0m ago	100
11	GPT-3.5 Turbo Instruct	OpenAI	#290	0m ago	100
12	Gemma 2 9B	Google	#291	0m ago	100
13	LlamaGuard 2 8B	Meta	#292	0m ago	100
14	Mistral 7B Instruct v0.1	Mistral AI	#293	0m ago	100
15	Mixtral 8x22B Instruct	Mistral AI	#279	0m ago	100
16	Qwen VL Plus	Alibaba	#202	0m ago	100
17	UI-TARS 7B	ByteDance	#203	0m ago	100
18	Cogito v2.1 671B	deepcogito	#204	0m ago	100
19	Voxtral Small 24B 2507	Mistral AI	#205	0m ago	100
20	ERNIE 4.5 21B A3B	Baidu	#206	0m ago	100

Stalest Models

Bottom 20

Models with the oldest data. These may have outdated scores or missing signal updates.

#	Model	Provider	Rank	Last Updated	Freshness
1	GPT-5.4 Pro	OpenAI	#1	0m ago	100
2	GPT-5.2 Pro	OpenAI	#2	0m ago	100
3	GPT-5 Pro	OpenAI	#3	0m ago	100
4	o3 Pro	OpenAI	#4	0m ago	100
5	Claude Opus 4.1	Anthropic	#5	0m ago	100
6	o1-pro	OpenAI	#6	0m ago	100
7	Claude Opus 4	Anthropic	#7	0m ago	100
8	o3 Deep Research	OpenAI	#8	0m ago	100
9	Claude Opus 4.6	Anthropic	#9	0m ago	100
10	Claude Opus 4.5	Anthropic	#10	0m ago	100
11	GPT-5.4	OpenAI	#11	0m ago	100
12	Claude Sonnet 4.5	Anthropic	#12	0m ago	100
13	Qwen3 VL 30B A3B Thinking	Alibaba	#13	0m ago	100
14	Qwen3 VL 235B A22B Thinking	Alibaba	#14	0m ago	100
15	GPT-5.2	OpenAI	#15	0m ago	100
16	Gemini 3.1 Pro Preview Custom Tools	Google	#16	0m ago	100
17	Gemini 3.1 Pro Preview	Google	#17	0m ago	100
18	Gemini 3 Pro Preview	Google	#18	0m ago	100
19	Claude Sonnet 4.6	Anthropic	#19	0m ago	100
20	GPT-5.1	OpenAI	#20	0m ago	100

Update Frequency by Provider

How fresh is each provider's data? Sorted by average freshness score across all their models.

Provider	Models	Avg Score	Freshest Model	Stalest Model
Inflection	2	100.0	Inflection 3 Pi	Inflection 3 Pi
Microsoft	2	100.0	WizardLM-2 8x22B	Phi 4
Meta	17	100.0	Llama 3 70B Instruct	Llama 4 Maverick
deepcogito	1	100.0	Cogito v2.1 671B	Cogito v2.1 671B
Cohere	4	100.0	Command A	Command A
essentialai	1	100.0	Rnj 1 Instruct	Rnj 1 Instruct
IBM	1	100.0	Granite 4.0 Micro	Granite 4.0 Micro
eleutherai	1	100.0	Llemma 7b	Llemma 7b
Liquid AI	5	100.0	LFM2.5-1.2B-Instruct (free)	LFM2.5-1.2B-Thinking (free)
aion-labs	3	100.0	Aion-1.0	Aion-2.0
Mistral AI	25	100.0	Mistral Large	Mistral Small 3.2 24B
Upstage	1	100.0	Solar Pro 3	Solar Pro 3
Writer	1	100.0	Palmyra X5	Palmyra X5
Amazon	5	100.0	Nova Pro 1.0	Nova 2 Lite
Perplexity	5	100.0	Sonar Deep Research	Sonar Pro Search
Baidu	5	100.0	ERNIE 4.5 21B A3B	ERNIE 4.5 VL 28B A3B
Allen AI	7	100.0	Olmo 3.1 32B Instruct	Olmo 3.1 32B Think
Moonshot AI	5	100.0	Kimi K2 0711	Kimi K2.5
arcee-ai	7	100.0	Virtuoso Large	Trinity Large Preview (free)
Inception	3	100.0	Mercury	Mercury 2
DeepSeek	12	100.0	DeepSeek V3	DeepSeek V3.2
MiniMax	6	100.0	MiniMax M2-her	MiniMax M2.5
Alibaba	51	100.0	Qwen2.5-VL 7B Instruct	Qwen3 VL 30B A3B Thinking
AI21 Labs	1	100.0	Jamba Large 1.7	Jamba Large 1.7
Tencent	1	100.0	Hunyuan A13B Instruct	Hunyuan A13B Instruct
NVIDIA	8	100.0	Llama 3.1 Nemotron 70B Instruct	Nemotron Nano 12B 2 VL (free)
ByteDance	4	100.0	UI-TARS 7B	Seed-2.0-Mini
OpenAI	59	100.0	GPT-3.5 Turbo (older v0613)	GPT-5.4 Pro
Google	24	100.0	Gemma 2 27B	Gemini 3.1 Pro Preview Custom Tools
xAI	8	100.0	Grok 3 Beta	Grok 4.1 Fast
StepFun	2	100.0	Step 3.5 Flash (free)	Step 3.5 Flash (free)
Xiaomi	1	100.0	MiMo-V2-Flash	MiMo-V2-Flash
Kuaishou	1	100.0	KAT-Coder-Pro V1	KAT-Coder-Pro V1
Meituan	1	100.0	LongCat Flash Chat	LongCat Flash Chat
Anthropic	13	100.0	Claude 3 Haiku	Claude Opus 4.1

Methodology

Freshness Score

Computed as 100 minus the number of hours since the last update, capped at 100 hours. A score of 100 means the model was just updated; a score of 0 means it has not been updated in over 100 hours (4+ days).

Freshness Tiers

Models are grouped into five tiers based on update recency: Just Updated (<1h), Fresh (1-6h), Recent (6-24h), Aging (1-7d), and Stale (>7d). Models without a known update timestamp are classified as Stale.

Why Freshness Matters

Stale data can lead to inaccurate rankings and outdated signal scores. Models with fresh data reflect the latest pricing, capabilities, and benchmark results. Monitoring freshness helps identify data pipeline issues early.