Leaderboard

coding rankings

Updated weekly

Each leaderboard uses transparent weighted scoring, current model context, and supporting analysis to help teams interpret the results with confidence. Only full-profile entries appear in rankings; broader catalog records remain available elsewhere on the site when only source-backed metadata is currently available.

Full-profile entries

245

Models with complete enough metadata and scoring coverage to be meaningfully ranked in this category.

Ranking basis

coding

Scores combine benchmark evidence, product metadata, and cost/context signals when those fields are published.

Catalog caveat

Tracked models without full scoring remain in the directory and provider pages, but are not relied on for analytical ranking claims.

Rank	Model	Provider	Score	Context
#1	GPT-5.4	OpenAI	95	1,000,000
#2	GPT-5.2	OpenAI	93	1,000,000
#3	GPT-5.3-Codex	OpenAI	93	1,000,000
#4	Claude Sonnet 4.5	Anthropic	92	1,000,000
#5	Claude Sonnet 4.6	Anthropic	92	1,000,000
#6	GPT-5.4 Pro	OpenAI	92	1,000,000
#7	Claude 3.7 Sonnet	Anthropic	91	200,000
#8	Claude Opus 4.6	Anthropic	91	1,000,000
#9	Gemini 3.1 Pro	Google DeepMind	91	1,048,576
#10	GPT-5.2 Pro	OpenAI	91	1,000,000
#11	GPT-4o	OpenAI	90	128,000
#12	GPT-5	OpenAI	90	1,000,000
#13	Claude Sonnet 4	Anthropic	89	1,000,000
#14	Gemini 2.5 Pro	Google DeepMind	89	1,048,576
#15	Gemini 3.0 Pro	Google DeepMind	89	1,048,576
#16	Gemini 3.1 Flash	Google DeepMind	89	1,048,576
#17	Claude Opus 4.1	Anthropic	88	200,000
#18	Gemini 2.5 Pro TTS	Google DeepMind	87	1,048,576
#19	Gemini 3.0 Flash	Google DeepMind	87	1,048,576
#20	GPT-4.1	OpenAI	87	1,048,576
#21	GPT-5 mini	OpenAI	87	1,000,000
#22	Mistral Large 25	Mistral AI	87	128,000
#23	Claude Opus 4	Anthropic	86	200,000
#24	Gemini 2.5 Flash	Google DeepMind	86	1,048,576
#25	GPT-5.3 Instant	OpenAI	86	1,000,000
#26	Llama 4 Maverick	Meta	85	1,048,576
#27	o4-mini	OpenAI	85	200,000
#28	Claude Haiku 4.5	Anthropic	84	200,000
#29	Command R+ 2026	Cohere	84	128,000
#30	morph-v3-fast-apply	Morph	84	128,000
#31	warpgrep-v2	Morph	84	128,000
#32	Gemini 2.0 Flash	Google DeepMind	83	1,048,576
#33	o3	OpenAI	83	200,000
#34	o4-mini-deep-research	OpenAI	82	200,000
#35	Gemini 2.5 Flash Live	Google DeepMind	81	1,048,576
#36	Gemini 2.5 Flash Native Audio Preview	Google DeepMind	81	1,048,576
#37	Gemini 2.5 Flash-Lite	Google DeepMind	81	1,048,576
#38	GPT-4.1 mini	OpenAI	81	1,048,576
#39	o3-deep-research	OpenAI	81	200,000
#40	flash-compact	Morph	80	200,000
#41	Gemini 1.5 Pro	Google DeepMind	80	2,097,152
#42	Sonar Reasoning Pro	Perplexity	80	200,000
#43	Gemini 3.1 Flash-Lite	Google DeepMind	79	1,048,576
#44	GPT-5 nano	OpenAI	79	1,000,000
#45	gpt-oss-120b	OpenAI	79	131,072
#46	Llama 3.3 70B Instruct	Meta	79	128,000
#47	Claude Haiku 3.5	Anthropic	78	200,000
#48	o1	OpenAI	78	200,000
#49	o1-mini	OpenAI	78	128,000
#50	Claude Opus 3	Anthropic	77	200,000
#51	Claude Sonnet 3	Anthropic	77	200,000
#52	GPT-4o-mini	OpenAI	77	128,000
#53	Nova Pro	Amazon Web Services	77	300,000
#54	Sonar Pro	Perplexity	77	200,000
#55	Gemini 1.5 Flash	Google DeepMind	76	1,048,576
#56	Gemini 2.0 Flash-Lite	Google DeepMind	76	1,048,576
#57	Step-3.5-Flash	StepFun	76	131,072
#58	Sonar Deep Research	Perplexity	75	200,000
#59	Llama 4 Scout	Meta	74	10,485,760
#60	Sonar	Perplexity	74	128,000
#61	Claude Haiku 3	Anthropic	73	200,000
#62	Claude Haiku 3.5	Anthropic	73	200,000
#63	Claude Haiku 4.5	Anthropic	73	200,000
#64	Codestral	Mistral AI	73	256,000
#65	Doubao-Seed-1.6	ByteDance / Doubao	73	128,000
#66	Doubao-Seed-1.6-Flash	ByteDance / Doubao	73	128,000
#67	Doubao-Seed-2.0-Code	ByteDance / Doubao	73	128,000
#68	Doubao-Seed-Code	ByteDance / Doubao	73	128,000
#69	gpt-audio	OpenAI	73	128,000
#70	gpt-realtime	OpenAI	73	128,000
#71	Llama 3.1 405B Instruct	Meta	73	128,000
#72	Llama 3.1 70B Instruct	Meta	73	128,000
#73	MiniMax-M1	MiniMax	73	204,800
#74	MiniMax-Text-01	MiniMax	73	204,800
#75	MiniMax-VL-01	MiniMax	73	204,800
#76	Voxtral Mini Transcribe	Mistral AI	73	131,072
#77	Codestral Embed	Mistral AI	72	32,768
#78	CogVideoX	Z.AI	72	128,000
#79	CogView 4	Z.AI	72	128,000
#80	Devstral Medium 1.0	Mistral AI	72	128,000
#81	ERNIE 3.5 128K	Baidu / ERNIE	72	128,000
#82	ERNIE 4.0 Turbo 8K	Baidu / ERNIE	72	128,000
#83	ERNIE 4.5 Turbo 32K	Baidu / ERNIE	72	32,768
#84	ERNIE Functions 8K	Baidu / ERNIE	72	128,000
#85	ERNIE Speed 128K	Baidu / ERNIE	72	128,000
#86	GLM-4.5	Z.AI	72	128,000
#87	GLM-4.5V	Z.AI	72	128,000
#88	GLM-4.6	Z.AI	72	128,000
#89	GLM-4.6V	Z.AI	72	128,000
#90	GLM-4.7	Z.AI	72	128,000
#91	GLM-5	Z.AI	72	128,000
#92	GLM-Image	Z.AI	72	128,000
#93	GLM-OCR	Z.AI	72	128,000
#94	Hunyuan Code	Tencent / Hunyuan	72	128,000
#95	Hunyuan Lite	Tencent / Hunyuan	72	128,000
#96	Hunyuan Standard	Tencent / Hunyuan	72	128,000
#97	Hunyuan T1	Tencent / Hunyuan	72	256,000
#98	Hunyuan T1 Vision	Tencent / Hunyuan	72	128,000
#99	Hunyuan TurboS	Tencent / Hunyuan	72	128,000
#100	Hunyuan TurboS LongText 128K	Tencent / Hunyuan	72	128,000
#101	Kimi K2	Moonshot AI / Kimi	72	131,072
#102	Kimi K2 Thinking	Moonshot AI / Kimi	72	256,000
#103	Kimi K2 Turbo Preview	Moonshot AI / Kimi	72	256,000
#104	Kimi K2.5	Moonshot AI / Kimi	72	256,000
#105	Magistral Medium 1.2	Mistral AI	72	128,000
#106	MiniMax-M2	MiniMax	72	204,800
#107	MiniMax-M2.1	MiniMax	72	204,800
#108	MiniMax-M2.1-highspeed	MiniMax	72	204,800
#109	MiniMax-M2.5	MiniMax	72	204,800
#110	MiniMax-M2.5-highspeed	MiniMax	72	204,800
#111	Mistral Embed	Mistral AI	72	32,768
#112	Mistral Large 3	Mistral AI	72	128,000
#113	Mistral Medium 3.1	Mistral AI	72	128,000
#114	Mistral Moderation	Mistral AI	72	32,768
#115	Mistral OCR 2505	Mistral AI	72	32,768
#116	Mistral Small 3.1	Mistral AI	72	128,000
#117	Mistral Small 3.2 Open	Mistral AI	72	128,000
#118	Pixtral 12B	Mistral AI	72	131,072
#119	Pixtral Large	Mistral AI	72	131,072
#120	Vidu Q1	Z.AI	72	128,000
#121	Voxtral Mini Open	Mistral AI	72	131,072
#122	Voxtral Small Open	Mistral AI	72	131,072
#123	Claude Haiku 3	Anthropic	71	200,000
#124	Claude Sonnet 3	Anthropic	71	200,000
#125	Claude Sonnet 4	Anthropic	71	1,000,000
#126	Command A	Cohere	71	256,000
#127	Command A Reasoning	Cohere	71	128,000
#128	Command A Translate	Cohere	71	128,000
#129	Command A Vision	Cohere	71	128,000
#130	Command R+	Cohere	71	128,000
#131	Command R7B	Cohere	71	128,000
#132	DeepSeek-Coder-V2	DeepSeek	71	128,000
#133	DeepSeek-Math-V2	DeepSeek	71	128,000
#134	DeepSeek-R1	DeepSeek	71	128,000
#135	DeepSeek-R1-Distill-Llama-70B	DeepSeek	71	128,000
#136	DeepSeek-V2.5	DeepSeek	71	128,000
#137	DeepSeek-V3	DeepSeek	71	128,000
#138	DeepSeek-V3.1	DeepSeek	71	128,000
#139	DeepSeek-V3.1-Base	DeepSeek	71	128,000
#140	DeepSeek-V3.2	DeepSeek	71	128,000
#141	DeepSeek-V3.2-Exp	DeepSeek	71	128,000
#142	Devstral 2 Open	Mistral AI	71	128,000
#143	Devstral Small 2	Mistral AI	71	128,000
#144	Embed 4	Cohere	71	128,000
#145	Grok 3	xAI	71	131,072
#146	Grok 3 Mini	xAI	71	131,072
#147	Grok 4	xAI	71	256,000
#148	Grok 4 Fast Reasoning	xAI	71	131,072
#149	grok-image	xAI	71	131,072
#150	image-01	MiniMax	71	8,192
#151	image-01-live	MiniMax	71	8,192
#152	Jamba 3B	AI21 Labs	71	256,000
#153	Jamba Large	AI21 Labs	71	256,000
#154	Jamba Large 1.6	AI21 Labs	71	256,000
#155	Jamba Mini	AI21 Labs	71	256,000
#156	Jamba Mini 1.6	AI21 Labs	71	256,000
#157	Jamba Mini 1.7	AI21 Labs	71	256,000
#158	Magistral Small 1.2 Open	Mistral AI	71	128,000
#159	MiniMax-Speech-02	MiniMax	71	8,192
#160	Ministral 3 14B Open	Mistral AI	71	128,000
#161	Ministral 3 3B Open	Mistral AI	71	128,000
#162	Ministral 3 8B Open	Mistral AI	71	128,000
#163	Mistral Large 3 Open	Mistral AI	71	128,000
#164	Mistral Nemo 12B	Mistral AI	71	128,000
#165	music-2.0	MiniMax	71	8,192
#166	Nova Lite	Amazon Web Services	71	300,000
#167	Phi-3-vision-128k-instruct	Microsoft	71	128,000
#168	Phi-3.5-mini-instruct	Microsoft	71	131,072
#169	Phi-3.5-MoE-instruct	Microsoft	71	131,072
#170	Phi-3.5-vision-instruct	Microsoft	71	131,072
#171	Phi-4-mini-flash-reasoning	Microsoft	71	131,072
#172	Phi-4-mini-instruct	Microsoft	71	131,072
#173	Phi-4-multimodal-instruct	Microsoft	71	131,072
#174	Phi-4-reasoning	Microsoft	71	131,072
#175	Phi-4-reasoning-plus	Microsoft	71	131,072
#176	Phi-4-reasoning-vision-15B	Microsoft	71	131,072
#177	Qwen2.5-1.5B-Instruct	Alibaba Qwen	71	131,072
#178	Qwen2.5-14B-Instruct	Alibaba Qwen	71	131,072
#179	Qwen2.5-32B-Instruct	Alibaba Qwen	71	131,072
#180	Qwen2.5-3B-Instruct	Alibaba Qwen	71	131,072
#181	Qwen2.5-72B-Instruct	Alibaba Qwen	71	131,072
#182	Qwen2.5-7B-Instruct	Alibaba Qwen	71	131,072
#183	Qwen2.5-Max	Alibaba Qwen	71	131,072
#184	Qwen2.5-VL-72B-Instruct	Alibaba Qwen	71	131,072
#185	Qwen2.5-VL-7B-Instruct	Alibaba Qwen	71	131,072
#186	Qwen3-Coder-Next	Alibaba Qwen	71	131,072
#187	Qwen3.5-0.8B	Alibaba Qwen	71	131,072
#188	Qwen3.5-122B-A10B	Alibaba Qwen	71	131,072
#189	Qwen3.5-27B	Alibaba Qwen	71	131,072
#190	Qwen3.5-2B	Alibaba Qwen	71	131,072
#191	Qwen3.5-35B-A3B	Alibaba Qwen	71	131,072
#192	Qwen3.5-397B-A17B	Alibaba Qwen	71	131,072
#193	Qwen3.5-4B	Alibaba Qwen	71	131,072
#194	Qwen3.5-9B	Alibaba Qwen	71	131,072
#195	DeepSeek-OCR	DeepSeek	70	16,384
#196	DeepSeek-OCR-2	DeepSeek	70	16,384
#197	DeepSeek-VL2-Small	DeepSeek	70	16,384
#198	Gemini 1.5 Flash-8B	Google DeepMind	70	1,048,576
#199	Janus-Pro-7B	DeepSeek	70	16,384
#200	Phi-4	Microsoft	70	16,384
#201	gpt-oss-20b	OpenAI	69	131,072
#202	phi-1	Microsoft	69	4,096
#203	phi-1_5	Microsoft	69	4,096
#204	phi-2	Microsoft	69	4,096
#205	Phi-3-medium-4k-instruct	Microsoft	69	4,096
#206	Phi-3-mini-4k-instruct	Microsoft	69	4,096
#207	Phi-tiny-MoE-instruct	Microsoft	69	4,096
#208	Claude Opus 3	Anthropic	68	200,000
#209	Claude Opus 4	Anthropic	68	200,000
#210	gpt-audio-mini	OpenAI	66	128,000
#211	gpt-realtime-mini	OpenAI	66	128,000
#212	Llama 3.1 8B Instruct	Meta	66	128,000
#213	Nova Micro	Amazon Web Services	66	128,000
#214	Code Llama 70B Instruct	Meta	65	8,192
#215	Llama 3.2 90B Vision Instruct	Meta	65	128,000
#216	Meta Llama 3 70B Instruct	Meta	65	8,192
#217	LFM2-24B-A2B	Liquid AI	62	32,768
#218	Step3-VL-10B	StepFun	62	131,072
#219	Code Llama 34B Instruct	Meta	61	8,192
#220	Llama 3.2 1B Instruct	Meta	61	128,000
#221	Llama 3.2 3B Instruct	Meta	61	128,000
#222	Meta Llama 3 8B Instruct	Meta	61	8,192
#223	Llama 3.2 11B Vision Instruct	Meta	60	128,000
#224	MiMo-VL-7B	Xiaomi	60	131,072
#225	MiMo-Audio-7B	Xiaomi	57	131,072
#226	LFM2-8B-A1B	Liquid AI	56	32,768
#227	Step-Audio-R1.1	StepFun	56	131,072
#228	GPT-4o mini Transcribe	OpenAI	55	128,000
#229	GPT-4o mini TTS	OpenAI	55	128,000
#230	GPT-4o Transcribe	OpenAI	55	128,000
#231	GPT Image 1	OpenAI	52	32,768
#232	chatgpt-image-latest	OpenAI	50	32,768
#233	gpt-image-1-mini	OpenAI	50	32,768
#234	LFM2.5-1.2B-Instruct	Liquid AI	48	131,072
#235	LFM2.5-1.2B-Thinking	Liquid AI	48	131,072
#236	LFM2-2.6B	Liquid AI	46	32,768
#237	Llama Guard 4 12B	Meta	45	131,072
#238	Llama Guard 3 11B Vision	Meta	42	131,072
#239	pplx-embed-v1-4b	Perplexity	35	8,192
#240	pplx-embed-v1-0.6b	Perplexity	34	8,192
#241	FLUX 1.1 Pro	Black Forest Labs	32	512
#242	FLUX 1 Pro	Black Forest Labs	31	512
#243	FLUX 1.1 Pro Ultra	Black Forest Labs	31	512
#244	Prompt Guard 86M	Meta	29	512
#245	NextStep-1.1	StepFun	23	512

Why #1: GPT-5.4

OpenAI's GPT-5.4, the most capable and efficient frontier model for professional work. First general-purpose model with native computer-use capabilities. Combines industry-leading coding from GPT-5.3-Codex with improved agentic workflows.

This model clears the current full-profile threshold for leaderboard methodology.

Why #2: GPT-5.2

OpenAI's latest frontier model with 1M token context for advanced reasoning and tool use.

This model clears the current full-profile threshold for leaderboard methodology.

Why #3: GPT-5.3-Codex

OpenAI's GPT-5.3-Codex for expanding Codex across the full spectrum of professional work on a computer. Powers the Codex app on macOS and Windows.

This model clears the current full-profile threshold for leaderboard methodology.