Degradation Tracker

Detect when AI models may be declining. This tracker monitors rank movements (position changes on the leaderboard) and flags models that have dropped significantly over 24 hours or 7 days. A higher “degradation points” number means more warning signs.

Models at Risk

280

Declining (7d)

223

Fragile

201

Sustained Decline

Top Models by Degradation Risk Score

LMMarketCap.com

Models at Risk

280 models showing signs of degradation, ranked by risk score. Higher risk scores indicate more concerning performance trends.

Deg. Pts	Model	Provider	Quality	Rank 24h	Rank 7d	Severity	State
85	Claude Opus 4.1Anthropic	Anthropic	74.4	-60	-10	high	fragile
37	Claude Sonnet 4Anthropic	Anthropic	73.9	-4	-14	high	fragile
27	GPT-4o (2024-08-06)OpenAI	OpenAI	71.2	+2	-11	high	fragile
27	GPT-4oOpenAI	OpenAI	71.2	+2	-11	high	fragile
27	GPT-4o (2024-05-13)OpenAI	OpenAI	71.2	+2	-11	high	fragile
27	MiniMax M2MiniMax	MiniMax	71.0	+2	-11	high	fragile
27	MiniMax M1MiniMax	MiniMax	70.8	+2	-11	high	fragile
27	GLM 4.5 AirZhipu AI	Zhipu AI	70.7	+2	-11	high	fragile
27	Llama 4 MaverickMeta	Meta	70.7	+2	-11	high	fragile
27	Qwen3 30B A3BAlibaba	Alibaba	64.1	+2	-11	high	fragile
26	Ling 3.0 Flash Sante (free)inclusionai	inclusionai	40.0	-1	-10	high	fragile
25	Qwen3.6 Max PreviewAlibaba	Alibaba	74.8	+1	-10	high	fragile
25	Hy3Tencent	Tencent	74.4	+1	-10	high	fragile
25	R1DeepSeek	DeepSeek	73.8	0	-10	high	fragile
25	MiniMax M2.5MiniMax	MiniMax	71.6	+2	-10	high	fragile
24	GPT-5 ProOpenAI	OpenAI	88.7	-1	-9	high	fragile
24	GPT-5 Pro (batch)OpenAI	OpenAI	88.7	-1	-9	high	fragile
24	GPT-5OpenAI	OpenAI	88.7	-1	-9	high	fragile
24	GPT-5 (batch)OpenAI	OpenAI	88.7	-1	-9	high	fragile
24	Gemini 3 Flash PreviewGoogle	Google	88.4	-1	-9	high	fragile
24	Gemini 3 Flash Preview (batch)Google	Google	88.4	-1	-9	high	fragile
24	Muse Spark 1.3 Contributormeta	meta	40.0	-1	-9	high	fragile
23	o1OpenAI	OpenAI	74.4	+1	-9	high	fragile
23	o3 Mini (batch)OpenAI	OpenAI	74.3	+1	-9	high	fragile
23	Qwen3.6 PlusAlibaba	Alibaba	74.1	+1	-9	high	fragile
23	MiniMax M3MiniMax	MiniMax	73.9	+1	-9	high	fragile
23	o1-proOpenAI	OpenAI	73.6	+1	-9	high	fragile
23	Qwen3.8 27BAlibaba	Alibaba	73.4	+1	-9	high	fragile
23	MiMo-V2.5Xiaomi	Xiaomi	73.0	+1	-9	high	fragile
23	Gemma 4 26B A4B Google	Google	73.0	+1	-9	high	fragile
23	Gemma 4 26B A4B (free)Google	Google	73.0	+1	-9	high	fragile
23	Qwen3.8 27B (free)Alibaba	Alibaba	72.9	+1	-9	high	fragile
23	Inklingthinkingmachines	thinkingmachines	72.9	+1	-9	high	fragile
23	Inkling (free)thinkingmachines	thinkingmachines	72.9	+1	-9	high	fragile
23	DeepSeek V3 0324DeepSeek	DeepSeek	71.8	+2	-9	high	fragile
23	Mistral Medium 3.5Mistral AI	Mistral AI	71.6	+2	-9	high	fragile
23	Mistral LargeMistral AI	Mistral AI	65.9	+2	-9	high	fragile
23	Composer 2Cursor	Cursor	65.7	+2	-9	high	fragile
23	Composer 2 FastCursor	Cursor	65.7	+2	-9	high	fragile
23	GLM 4.6VZhipu AI	Zhipu AI	65.6	+2	-9	high	fragile
23	Qwen3 235B A22B Thinking 2507Alibaba	Alibaba	65.3	+2	-9	high	fragile
23	Llama 3.1 70B InstructMeta	Meta	65.3	+2	-9	high	fragile
23	GPT-4OpenAI	OpenAI	64.9	+2	-9	high	fragile
23	Qwen3 235B A22B Instruct 2507Alibaba	Alibaba	64.7	+2	-9	high	fragile
23	o3 Mini HighOpenAI	OpenAI	63.9	+2	-9	high	fragile
23	GLM 4.7 FlashZhipu AI	Zhipu AI	63.5	+2	-9	high	fragile
23	Mixtral 8x22B InstructMistral AI	Mistral AI	63.4	+2	-9	high	fragile
23	Trinity Large Thinkingarcee-ai	arcee-ai	63.1	+2	-9	high	fragile
23	GPT-4o-mini (batch)OpenAI	OpenAI	62.5	+2	-9	high	fragile
23	GLM 4.5VZhipu AI	Zhipu AI	62.3	+2	-9	high	fragile
23	GPT-5 NanoOpenAI	OpenAI	61.9	+2	-9	high	fragile
23	gpt-oss-120bOpenAI	OpenAI	61.6	+2	-9	high	fragile
23	GPT-4o-mini (2024-07-18)OpenAI	OpenAI	56.5	+2	-9	high	fragile
23	GPT-4.1 MiniOpenAI	OpenAI	56.2	+2	-9	high	fragile
23	Ternary Bonsai 2 27Bprism-ml	prism-ml	40.0	0	-9	high	fragile
23	Paretounbiased	unbiased	40.0	0	-9	high	fragile
23	DeepSeek Pro Latest~deepseek	~deepseek	40.0	0	-9	high	fragile
23	DeepSeek Flash Latest~deepseek	~deepseek	40.0	0	-9	high	fragile
23	Schematron V2 Turboinference-net	inference-net	40.0	0	-9	high	fragile
23	Schematron V2 Smallinference-net	inference-net	40.0	0	-9	high	fragile
23	GPT Astra Latest~openai	~openai	40.0	0	-9	high	fragile
23	GPT Sol Latest~openai	~openai	40.0	0	-9	high	fragile
23	GPT Terra Latest~openai	~openai	40.0	0	-9	high	fragile
23	GPT Luna Latest~openai	~openai	40.0	0	-9	high	fragile
23	Fugu Ultra v2sakana	sakana	40.0	0	-9	high	fragile
23	Fugu Maxsakana	sakana	40.0	0	-9	high	fragile
23	Ling 3.0 Flash VLinclusionai	inclusionai	40.0	0	-9	high	fragile
23	Ling 3.0 Flash VL (free)inclusionai	inclusionai	40.0	0	-9	high	fragile
23	DeepSeek V4.1 FlashDeepSeek	DeepSeek	40.0	0	-9	high	fragile
22	Grok 4.20 Multi-AgentxAI	xAI	87.9	-1	-8	high	fragile
22	Hy4 previewTencent	Tencent	40.0	-1	-8	high	fragile
22	Ling 3.0 Flash Fininclusionai	inclusionai	40.0	-1	-8	high	fragile
22	Ling 3.0 Flash Fin (free)inclusionai	inclusionai	40.0	-1	-8	high	fragile
22	GLM Flash Latest~z-ai	~z-ai	40.0	-1	-8	high	fragile
22	Qwen3.8 FlashAlibaba	Alibaba	40.0	-1	-8	high	fragile
22	Muse Spark 1.2 Contributormeta	meta	40.0	-1	-8	high	fragile
22	DeepSeek V4 Flash Vision ExpDeepSeek	DeepSeek	40.0	-1	-8	high	fragile
21	GLM 5Zhipu AI	Zhipu AI	78.0	+2	-8	high	fragile
21	GLM 5.3 FlashXZhipu AI	Zhipu AI	77.9	+2	-8	high	fragile
21	GLM 5.3 Flash (batch)Zhipu AI	Zhipu AI	77.9	+2	-8	high	fragile
21	GLM 4.5Zhipu AI	Zhipu AI	75.1	+1	-8	high	fragile
21	Claude Haiku 4.5Anthropic	Anthropic	69.9	+2	-8	high	fragile
21	Claude Haiku 4.5 (batch)Anthropic	Anthropic	69.9	+2	-8	high	fragile
21	Qwen3 Max ThinkingAlibaba	Alibaba	68.2	+2	-8	high	fragile
21	MiniMax M2-herMiniMax	MiniMax	68.1	+2	-8	high	fragile
21	GPT-4.1OpenAI	OpenAI	67.7	+2	-8	high	fragile
21	GPT-4.1 (batch)OpenAI	OpenAI	67.7	+2	-8	high	fragile
21	Qwen3 Next 80B A3B InstructAlibaba	Alibaba	66.7	+2	-8	high	fragile
21	GPT-4 TurboOpenAI	OpenAI	66.7	+2	-8	high	fragile
21	Qwen3.5-9BAlibaba	Alibaba	66.5	+2	-8	high	fragile
21	Step 3.5 FlashStepFun	StepFun	66.1	+2	-8	high	fragile
21	Qwen3 30B A3B Thinking 2507Alibaba	Alibaba	64.1	+2	-8	high	fragile
21	GPT-4 Turbo (batch)OpenAI	OpenAI	61.5	+3	-8	high	fragile
21	Mercury 2.5Inception	Inception	61.2	+3	-8	high	fragile
21	Qwen3 8BAlibaba	Alibaba	61.0	+3	-8	high	fragile
21	Mercury 2Inception	Inception	60.9	+3	-8	high	fragile
21	Nova 2 LiteAmazon	Amazon	60.4	+3	-8	high	fragile
21	Llama 4 ScoutMeta	Meta	60.2	+3	-8	high	fragile
21	Phi 4Microsoft	Microsoft	60.2	+3	-8	high	fragile
21	GPT-4.1 Mini (batch)OpenAI	OpenAI	58.9	+3	-8	high	fragile
21	GPT-4.1 Nano (batch)OpenAI	OpenAI	58.9	+3	-8	high	fragile
21	gpt-oss-20bOpenAI	OpenAI	57.4	+3	-8	high	fragile
21	Qwen3 235B A22BAlibaba	Alibaba	54.0	+2	-8	high	fragile
21	Granite 4.2 8BIBM	IBM	53.8	+2	-8	high	fragile
21	GPT-5 MiniOpenAI	OpenAI	53.8	+2	-8	high	fragile
20	Grok 4.20xAI	xAI	88.8	-1	-7	high	fragile
19	GLM 5.3 (batch)Zhipu AI	Zhipu AI	78.6	+1	-7	high	fragile
19	GLM 5.2Zhipu AI	Zhipu AI	78.6	+1	-7	high	fragile
19	Qwen3.5-122B-A10BAlibaba	Alibaba	77.7	+2	-7	high	fragile
19	Gemma 2 27BGoogle	Google	77.4	+2	-7	high	fragile
19	Qwen3.7 PlusAlibaba	Alibaba	75.6	+1	-7	high	fragile
19	GLM 5V TurboZhipu AI	Zhipu AI	72.3	+2	-7	high	fragile
19	o4 Mini HighOpenAI	OpenAI	72.1	+2	-7	high	fragile
19	GPT-4o (batch)OpenAI	OpenAI	72.0	+2	-7	high	fragile
19	DeepSeek V3DeepSeek	DeepSeek	69.5	+2	-7	high	fragile
19	Qwen3 VL 235B A22B InstructAlibaba	Alibaba	69.3	+2	-7	high	fragile
19	DeepSeek V3.1 TerminusDeepSeek	DeepSeek	69.3	+2	-7	high	fragile
19	GPT-4o-miniOpenAI	OpenAI	69.3	+2	-7	high	fragile
19	Inkling Smallthinkingmachines	thinkingmachines	68.7	+2	-7	high	fragile
19	Inkling Small (free)thinkingmachines	thinkingmachines	68.7	+2	-7	high	fragile
19	Qwen3.5-FlashAlibaba	Alibaba	68.6	+2	-7	high	fragile
19	Hy3 previewTencent	Tencent	68.4	+2	-7	high	fragile
19	Qwen3 MaxAlibaba	Alibaba	67.4	+2	-7	high	fragile
19	Mistral Large 3 2512 (batch)Mistral AI	Mistral AI	67.0	+2	-7	high	fragile
19	Llama 3.3 70B InstructMeta	Meta	66.8	+2	-7	high	fragile
19	Claude 3 HaikuAnthropic	Anthropic	51.3	+2	-7	high	fragile
19	Command ACohere	Cohere	50.8	+2	-7	high	fragile
19	Command R+ (08-2024)Cohere	Cohere	48.7	+2	-7	high	fragile
19	Kimi K2.5Moonshot AI	Moonshot AI	47.5	+2	-7	high	fragile
19	Llama 3.1 8B InstructMeta	Meta	44.5	+2	-7	high	fragile
19	GPT-4.1 NanoOpenAI	OpenAI	42.1	+2	-7	high	fragile
19	R1 Distill Llama 70BDeepSeek	DeepSeek	40.7	+2	-7	high	fragile
19	Hy-MT2-1.8BTencent	Tencent	40.0	0	-7	high	fragile
19	Hy-MT2-30B-A3BTencent	Tencent	40.0	0	-7	high	fragile
19	GLM Latest~z-ai	~z-ai	40.0	0	-7	high	fragile
19	Hy-MT2-7BTencent	Tencent	40.0	0	-7	high	fragile
19	Dots3-Note Preview (free)dots-studio	dots-studio	40.0	0	-7	high	fragile
19	Gemini 3.7 FlashGoogle	Google	40.0	0	-7	high	fragile
19	Gemini 3.7 Flash (batch)Google	Google	40.0	0	-7	high	fragile
19	Seed 2.1 TurboByteDance	ByteDance	40.0	0	-7	high	fragile
19	Qwen3.8 2.4T A95BAlibaba	Alibaba	40.0	0	-7	high	fragile
18	GPT-5.1 (batch)OpenAI	OpenAI	87.8	-1	-6	high	fragile
18	Claude Sonnet 4.6Anthropic	Anthropic	85.2	-1	-6	high	fragile
18	Claude Sonnet 4.6 (batch)Anthropic	Anthropic	85.2	-1	-6	high	fragile
18	Claude Opus 4.5Anthropic	Anthropic	85.1	-1	-6	high	fragile
18	Claude Opus 4.5 (batch)Anthropic	Anthropic	85.1	-1	-6	high	fragile
18	Gemini 2.5 ProGoogle	Google	83.5	-1	-6	high	fragile
18	Gemini 2.5 Pro (batch)Google	Google	83.5	-1	-6	high	fragile
18	Gemini 2.5 Pro Preview 06-05Google	Google	83.5	-1	-6	high	fragile
18	DeepSeek V3.2DeepSeek	DeepSeek	83.4	-1	-6	high	fragile
18	Claude Sonnet 4.5Anthropic	Anthropic	82.4	-1	-6	high	fragile
18	Claude Sonnet 4.5 (batch)Anthropic	Anthropic	82.4	-1	-6	high	fragile
17	Muse Spark 1.1meta	meta	80.9	+1	-6	high	fragile
17	Gemma 4 31BGoogle	Google	80.5	+1	-6	high	fragile
17	Gemma 4 31B (free)Google	Google	80.5	+1	-6	high	fragile
17	Qwen3.5 397B A17BAlibaba	Alibaba	79.4	+1	-6	high	fragile
17	R1 0528DeepSeek	DeepSeek	79.4	+1	-6	high	fragile
17	GPT-5.4 NanoOpenAI	OpenAI	79.3	+1	-6	high	fragile
17	GPT-5.4 Nano (batch)OpenAI	OpenAI	79.3	+1	-6	high	fragile
17	GPT-5.4 MiniOpenAI	OpenAI	79.3	+1	-6	high	fragile
17	GPT-5.4 Mini (batch)OpenAI	OpenAI	79.3	+1	-6	high	fragile
17	Gemini 3.1 Flash Lite PreviewGoogle	Google	79.3	+1	-6	high	fragile
17	Gemini 2.5 Flash LiteGoogle	Google	79.1	+1	-6	high	fragile
17	Gemini 2.5 Flash Lite (batch)Google	Google	79.1	+1	-6	high	fragile
17	Gemini 2.5 FlashGoogle	Google	79.1	+1	-6	high	fragile
17	Gemini 2.5 Flash (batch)Google	Google	79.1	+1	-6	high	fragile
17	Gemini 3.5 FlashGoogle	Google	79.0	+1	-6	high	fragile
17	Gemini 3.5 Flash (batch)Google	Google	79.0	+1	-6	high	fragile
17	GLM 5.3 FlashZhipu AI	Zhipu AI	78.0	+2	-6	high	fragile
17	MiMo-V2.5-ProXiaomi	Xiaomi	76.2	+1	-6	high	fragile
17	Qwen3.5-35B-A3BAlibaba	Alibaba	76.0	+1	-6	high	fragile
17	GLM 5.2 (free)Zhipu AI	Zhipu AI	75.7	+1	-6	high	fragile
17	Kimi K2.6Moonshot AI	Moonshot AI	75.7	+1	-6	high	fragile
17	o3 MiniOpenAI	OpenAI	75.3	+1	-6	high	fragile
17	Claude Opus 4.1 (batch)Anthropic	Anthropic	75.2	+1	-6	high	fragile
17	Command R (08-2024)Cohere	Cohere	48.7	+3	-6	high	fragile
17	Seed-2.0-CodeByteDance	ByteDance	40.0	+1	-6	high	fragile
17	DeepSeek V4 Pro 0813DeepSeek	DeepSeek	40.0	+1	-6	high	fragile
11	GPT-5.1-Codex-MiniOpenAI	OpenAI	87.8	-1	-5	high	stable
11	o3 ProOpenAI	OpenAI	86.7	-1	-5	high	stable
11	o3OpenAI	OpenAI	86.7	-1	-5	high	stable
11	o3 (batch)OpenAI	OpenAI	86.7	-1	-5	high	stable
10	GPT-6 AstraOpenAI	OpenAI	81.8	+1	-5	high	stable
10	GPT-6 Astra (batch)OpenAI	OpenAI	81.8	+1	-5	high	stable
10	GPT-6 Astra ProOpenAI	OpenAI	81.8	+1	-5	high	stable
10	GPT-6 Astra Pro (batch)OpenAI	OpenAI	81.8	+1	-5	high	stable
10	o4 MiniOpenAI	OpenAI	81.4	+1	-5	high	stable
10	o4 Mini (batch)OpenAI	OpenAI	81.4	+1	-5	high	stable
10	Gemini 3.8 FlashGoogle	Google	81.0	+1	-5	high	stable
10	Gemini 3.8 Flash (batch)Google	Google	81.0	+1	-5	high	stable
10	Qwen3.5-27BAlibaba	Alibaba	77.0	+2	-5	high	stable
10	GPT-5 Mini (batch)OpenAI	OpenAI	76.8	+2	-5	high	stable
10	GPT-5 Nano (batch)OpenAI	OpenAI	76.8	+2	-5	high	stable
10	Gemini 3.5 Flash LiteGoogle	Google	76.5	+2	-5	high	stable
10	Gemini 3.5 Flash Lite (batch)Google	Google	76.5	+2	-5	high	stable
10	LFM2.5-2.6B (free)Liquid AI	Liquid AI	40.0	+2	-5	high	stable
10	Nemotron 3.5 LightningNVIDIA	NVIDIA	40.0	+2	-5	high	stable
10	Nemotron 3.5 Lightning (free)NVIDIA	NVIDIA	40.0	+2	-5	high	stable
10	Sakana Namazusakana	sakana	40.0	+2	-5	high	stable
10	Solar Pro 4Upstage	Upstage	40.0	+2	-5	high	stable
10	Muse Glimmer 30Bmeta	meta	40.0	+2	-5	high	stable
8	Grok 4.3 (batch)xAI	xAI	81.0	+1	-4	medium	stable
8	MiniMax M2.1MiniMax	MiniMax	71.0	+2	-4	medium	stable
6	GPT-5.2 ProOpenAI	OpenAI	90.5	0	-3	medium	stable
6	GPT-5.2 Pro (batch)OpenAI	OpenAI	90.5	0	-3	medium	stable
6	GPT-5.2OpenAI	OpenAI	90.5	0	-3	medium	stable
6	GPT-5.2 (batch)OpenAI	OpenAI	90.5	0	-3	medium	stable
6	Claude Opus 4.6Anthropic	Anthropic	90.4	0	-3	medium	stable
6	Claude Opus 4.6 (batch)Anthropic	Anthropic	90.4	0	-3	medium	stable
6	GPT-5.6 Luna ProOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Luna Pro (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 LunaOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Luna (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Terra ProOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Terra Pro (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 TerraOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Terra (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Sol ProOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Sol Pro (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 SolOpenAI	OpenAI	89.0	0	-3	medium	stable
6	GPT-5.6 Sol (batch)OpenAI	OpenAI	89.0	0	-3	medium	stable
6	Grok 4.5xAI	xAI	88.8	-1	+57	medium	fragile
6	Grok 4.3xAI	xAI	88.8	-1	+31	medium	fragile
6	Claude Sonnet 5Anthropic	Anthropic	85.2	-1	+232	medium	fragile
6	DeepSeek V4 Flash Latest~deepseek	~deepseek	40.0	+3	-3	medium	stable
6	DeepSeek V4 Flash 0731DeepSeek	DeepSeek	40.0	+3	-3	medium	stable
5	Claude Opus 5Anthropic	Anthropic	95.1	0	+273	medium	fragile
5	GPT-5.5 ProOpenAI	OpenAI	92.7	0	+9	medium	fragile
5	GPT-5.3-CodexOpenAI	OpenAI	90.5	0	+19	medium	fragile
5	GPT-5.2 ChatOpenAI	OpenAI	90.5	0	+81	medium	fragile
5	Muse Spark 1.3meta	meta	80.9	+1	+162	medium	fragile
5	Muse Spark 1.2meta	meta	80.9	+1	+190	medium	fragile
5	GLM 5.1Zhipu AI	Zhipu AI	78.0	+2	+12	medium	fragile
5	GLM 5 TurboZhipu AI	Zhipu AI	78.0	+2	+44	medium	fragile
5	Qwen3.8 Max (0902)Alibaba	Alibaba	75.6	+1	+121	medium	fragile
5	GLM 4.7Zhipu AI	Zhipu AI	75.1	+1	+14	medium	fragile
5	GLM 4.6Zhipu AI	Zhipu AI	75.1	+1	+28	medium	fragile
5	Qwen3.7 MaxAlibaba	Alibaba	74.8	+1	+177	medium	fragile
5	Qwen3.5 Plus 2026-04-20Alibaba	Alibaba	74.8	+1	+188	medium	fragile
5	DeepSeek V3.1DeepSeek	DeepSeek	71.8	+2	+8	medium	fragile
5	MiniMax M2.7MiniMax	MiniMax	71.6	+2	+7	medium	fragile
5	GPT-4o (2024-11-20)OpenAI	OpenAI	71.2	+2	+61	medium	fragile
5	Qwen3.5 Plus 2026-02-15Alibaba	Alibaba	68.2	+2	+160	medium	fragile
5	Qwen3 Next 80B A3B ThinkingAlibaba	Alibaba	66.7	+2	+7	medium	fragile
5	Mistral Large 2407Mistral AI	Mistral AI	65.9	+2	+24	medium	fragile
5	Qwen3 30B A3B Instruct 2507Alibaba	Alibaba	64.1	+2	+171	medium	fragile
4	Claude Opus 4.7Anthropic	Anthropic	95.1	0	-2	low	stable
4	Claude Opus 4.7 (batch)Anthropic	Anthropic	95.1	0	-2	low	stable
4	GPT-5.5 Pro (batch)OpenAI	OpenAI	92.7	0	-2	low	stable
4	GPT-5.5OpenAI	OpenAI	92.7	0	-2	low	stable
4	GPT-5.5 (batch)OpenAI	OpenAI	92.7	0	-2	low	stable
4	Gemini 3.1 Pro Preview Custom ToolsGoogle	Google	92.2	0	-2	low	stable
4	Gemini 3.1 Pro PreviewGoogle	Google	92.2	0	-2	low	stable
4	Gemini 3.1 Pro Preview (batch)Google	Google	92.2	0	-2	low	stable
4	GPT-5.4 ProOpenAI	OpenAI	91.9	0	-2	low	stable
4	GPT-5.4 Pro (batch)OpenAI	OpenAI	91.9	0	-2	low	stable
4	GPT-5.4OpenAI	OpenAI	91.9	0	-2	low	stable
4	GPT-5.4 (batch)OpenAI	OpenAI	91.9	0	-2	low	stable
4	GPT-5.2-CodexOpenAI	OpenAI	90.5	0	-2	low	stable
4	Qwen3.7 FlashAlibaba	Alibaba	40.0	+4	-2	low	stable
2	Claude Opus 4.8 (batch)Anthropic	Anthropic	94.6	0	-1	low	stable
2	Claude Opus 5 (batch)Anthropic	Anthropic	40.0	+4	-1	low	stable
2	Ling 3.0 Flashinclusionai	inclusionai	40.0	+4	-1	low	stable
2	Laguna S 2.1poolside	poolside	40.0	+4	-1	low	stable
2	Laguna S 2.1 (free)poolside	poolside	40.0	+4	-1	low	stable
2	Gemini 3.6 FlashGoogle	Google	40.0	+4	-1	low	stable
2	Gemini 3.6 Flash (batch)Google	Google	40.0	+4	-1	low	stable
2	LongCat 2.0Meituan	Meituan	40.0	+4	-1	low	stable
2	Kimi K3Moonshot AI	Moonshot AI	40.0	+4	-1	low	stable
2	Kimi K3 (batch)Moonshot AI	Moonshot AI	40.0	+4	-1	low	stable
2	KAT-Coder-Pro V2.5Kuaishou	Kuaishou	40.0	+4	-1	low	stable
2	Grok Latest~x-ai	~x-ai	40.0	+4	-1	low	stable
2	Aion-3.0-Miniaion-labs	aion-labs	40.0	+4	-1	low	stable
2	Aion-3.0aion-labs	aion-labs	40.0	+4	-1	low	stable
2	Laguna XS 2.1poolside	poolside	40.0	+4	-1	low	stable
2	Laguna XS 2.1 (free)poolside	poolside	40.0	+4	-1	low	stable
1	Grok 4.6xAI	xAI	88.8	-1	+4	low	stable
1	GPT-5.1-Codex-MaxOpenAI	OpenAI	88.7	-1	+2	low	stable
1	GPT-5.1OpenAI	OpenAI	88.7	-1	+2	low	stable
1	GPT-5.1-CodexOpenAI	OpenAI	88.7	-1	+3	low	stable

Stable Models

14 models with no decline and a stable ranking state. These models are performing consistently.

#	Model	Provider	Score	24h	7d	State
1	Claude Fable 5.1Anthropic	Anthropic	95.9	0	0	stable
2	Claude Fable 5.1 (batch)Anthropic	Anthropic	95.9	0	0	stable
3	Claude Fable 5Anthropic	Anthropic	95.9	0	0	stable
4	Claude Fable 5 (batch)Anthropic	Anthropic	95.9	0	0	stable
6	Claude Opus 4.8Anthropic	Anthropic	95.1	0	+1	stable
100	GLM 5.3Zhipu AI	Zhipu AI	78.6	+1	+5	stable
150	DeepSeek V3.2 ExpDeepSeek	DeepSeek	71.8	+2	+5	stable
168	Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	69.3	+2	+5	stable
295	Claude Sonnet 5 (batch)Anthropic	Anthropic	40.0	+4	0	stable
296	Fugu Ultrasakana	sakana	40.0	+4	0	stable
297	North Mini Code (free)Cohere	Cohere	40.0	+4	0	stable
298	Kimi K2.7 CodeMoonshot AI	Moonshot AI	40.0	+4	0	stable
299	Claude Fable Latest~anthropic	~anthropic	40.0	+4	0	stable
300	Nemotron 3.5 Content SafetyNVIDIA	NVIDIA	40.0	+4	0	stable

How Degradation Is Detected

Our degradation detection system uses multiple signals to identify models that may be declining in quality or reliability.

Declining (7d)

Models whose 7-day rank change is worse than -2 positions. A sustained drop of more than two ranks over a week suggests the model may be losing ground to competitors or experiencing performance issues.

Fragile State

Models classified as "fragile" by our scoring system. These models have inconsistent performance metrics or borderline scores that could shift significantly with small changes in evaluation data.

Sustained Decline

Models declining on both the 24-hour and 7-day timeframes. When a model is losing rank on both short and medium-term windows, it indicates a persistent downward trend rather than temporary fluctuation.

Risk Score

The degradation risk score combines multiple signals: 7-day rank decline weighted 2x, 24-hour rank decline weighted 1x, plus 5 bonus points for fragile state. Higher scores indicate greater risk of meaningful performance degradation.

All Trackers

Coding, image, and video model trackers

Coding Tracker

Daily coding model performance and rankings

Leaderboard

Full model leaderboard with composite scores

Frequently Asked Questions

The tracker uses a multi-signal approach: it monitors 7-day rank decline (weighted 2x), 24-hour rank drops (weighted 1x), and fragile state classification (+5 points). Models are scored on a degradation risk scale where higher values indicate more warning signs of performance decline.

A fragile state indicates that a model has inconsistent performance metrics or borderline scores that could shift significantly with small changes in evaluation data. Fragile models are at higher risk of further ranking drops and warrant closer monitoring.

Yes, models can recover. Degradation may be temporary due to API issues, benchmark fluctuations, or scoring recalibrations. Models that show sustained decline over multiple weeks are more concerning than those with short-term dips. The tracker monitors both 24-hour and 7-day windows to help distinguish temporary noise from real trends.

Degradation Tracker

Top Models by Degradation Risk Score

Models at Risk

Stable Models

How Degradation Is Detected

Declining (7d)

Fragile State

Sustained Decline

Risk Score

Related