AI Model Rankings

Ranked by an AI Value score out of 100 — your chosen benchmarks, at your chosen weights, normalised against every model we track. Pick a preset or set the weights yourself; the ranking recalculates as you go.

Data through 14 Sept 2026.Every figure carries its source: 21% refresh automatically from a live API, the rest are published results we cite but have not re-measured ourselves. How each benchmark works

Metrics & Weights

6 metrics selected — total weight: 100%

Preset

Active metrics

SWE-bench Verified 30%HumanEval+ 15%Chatbot Arena ELO 15%Output Speed 15%Input Cost 13%Output Cost 13%

#	Model		30%	15%	15%	15%	12.5%	12.5%
1	Gemini 3 Flash Google	82.6	78.0%	86.0%	1473	200	$0.50	$3.0
2	GPT-5.2 OpenAI	81.2	80.0%	95.0%	1475	90	$1.8	$14.0
3	Gemini 3.1 Pro Google	80.6	80.6%	93.0%	1501	65	$2.0	$12.0
4	Claude Opus 4.6 Anthropic	80.2	80.8%	93.5%	1496	68	$5.0	$25.0
5	GPT-5.1 OpenAI	79.1	76.3%	93.0%	1464	95	$1.3	$10.0
6	Claude Opus 4.5 Anthropic	77.8	80.9%	92.0%	1468	50	$5.0	$25.0
7	GPT-5.1 Codex OpenAI	77.7	78.0%	96.0%	1395	85	$1.3	$10.0
8	GPT-5 OpenAI	75.7	74.9%	92.0%	1390	100	$1.3	$10.0
9	Claude Sonnet 4.6 Anthropic	74.9	79.6%	91.0%	1395	57	$3.0	$15.0
10	Gemini 3 Pro Google	73.7	65.0%	91.0%	1492	60	$2.0	$12.0
11	Grok 4.1 Fast xAI	73.7	60.0%	88.0%	1482	120	$0.20	$0.50
12	Claude Sonnet 4.5 Anthropic	73.6	77.2%	90.0%	1380	67	$3.0	$15.0
13	Grok 4 xAI	73.1	72.0%	90.0%	1430	55	$3.0	$15.0
14	Gemini 2.5 Pro Google	71.4	63.8%	89.0%	1465	55	$1.3	$10.0
15	DeepSeek V3.2 DeepSeek	71.3	73.0%	87.0%	1370	70	$0.27	$0.40
16	o4 Mini OpenAI	71.3	68.1%	88.0%	1350	120	$1.1	$4.4
17	Claude Sonnet 4 Anthropic	71	72.7%	88.0%	1365	75	$3.0	$15.0
18	Qwen 3.5 397B Qwen	70.8	76.4%	88.0%	1350	45	$0.55	$3.5
19	o3 OpenAI	69.7	69.1%	90.0%	1380	40	$2.0	$8.0
20	Claude Opus 4 Anthropic	68.7	72.5%	90.0%	1375	50	$15.0	$75.0
21	o3 Pro OpenAI	68.2	70.0%	92.0%	1410	25	$20.0	$80.0
22	GPT-5.1 Codex Mini OpenAI	67.6	55.0%	88.0%	1310	170	$0.25	$2.0
23	Qwen 3 Coder Qwen	67.5	66.5%	90.0%	1290	80	$0.30	$1.0
24	Claude 3.7 Sonnet Anthropic	65.5	62.3%	86.0%	1340	70	$3.0	$15.0
25	GPT-5 Mini OpenAI	64.2	48.0%	85.0%	1300	180	$0.25	$2.0
26	DeepSeek R1 0528 DeepSeek	64	55.0%	88.0%	1375	40	$0.50	$2.1
27	Grok 4 Fast xAI	63.6	48.0%	85.0%	1370	110	$0.20	$0.50
28	Gemini 2.5 Flash Google	63.6	38.0%	82.0%	1320	251	$0.30	$2.5
29	o3 Mini OpenAI	62.5	50.0%	87.0%	1320	100	$1.1	$4.4
30	GPT-4.1 OpenAI	62.4	50.0%	89.0%	1340	70	$2.0	$8.0
31	Mistral Large 25.12 Mistral	60.3	42.0%	82.0%	1418	70	$0.50	$1.5
32	DeepSeek R1 DeepSeek	60.2	49.2%	86.0%	1355	35	$0.70	$2.5
33	o1 Mini OpenAI	59.4	45.0%	89.0%	1300	80	$3.0	$12.0
34	Qwen 3 Max Qwen	59.4	46.0%	86.0%	1340	55	$0.78	$3.9
35	DeepSeek V3.1 DeepSeek	59.3	46.0%	84.0%	1340	65	$0.25	$0.95
36	Grok 3 xAI	59	42.0%	82.0%	1402	65	$3.0	$15.0
37	o1 OpenAI	59	48.9%	89.0%	1360	35	$15.0	$60.0
38	Qwen 3 235B Qwen	56.8	42.0%	84.0%	1320	60	$0.46	$1.8
39	Claude 3.5 Sonnet Anthropic	56.4	49.0%	81.7%	1268	70	$3.0	$15.0
40	GPT-4o OpenAI	56	30.7%	87.2%	1280	143	$2.5	$10.0
41	Codestral Mistral	56	40.0%	86.0%	1260	90	$0.30	$0.90
42	Claude Fable 5 Anthropic	55	95.0%	—%	—	60	$10.0	$50.0
43	GPT-4.1 Mini OpenAI	54.9	35.0%	80.0%	1250	160	$0.40	$1.6
44	Gemini 2.0 Flash Google	53.4	28.0%	76.0%	1240	220	$0.10	$0.40
45	Llama 4 Maverick Meta	53.3	35.0%	80.0%	1280	100	$0.20	$0.70
46	Claude Haiku 4.5 Anthropic	52.1	30.0%	78.0%	1220	180	$1.0	$5.0
47	Grok 3 Mini xAI	51.4	30.0%	78.0%	1260	130	$0.30	$0.50
48	Qwen 3 32B Qwen	50.9	30.0%	79.0%	1250	120	$0.08	$0.28
49	GPT-5 Nano OpenAI	50.8	25.0%	72.0%	1200	250	$0.05	$0.40
50	Pixtral Large Mistral	50.7	35.0%	80.0%	1270	60	$2.0	$6.0
51	Gemini 2.5 Flash Lite Google	49.7	22.0%	70.0%	1230	240	$0.10	$0.40
52	GPT-4o Mini OpenAI	49.6	20.0%	78.0%	1220	200	$0.15	$0.60
53	Llama 4 Scout Meta	49.3	28.0%	75.0%	1240	140	$0.10	$0.30
54	Mistral Medium 3.1 Mistral	49	28.0%	77.0%	1250	110	$0.40	$2.0
55	Command A Cohere	48.5	32.0%	78.0%	1250	70	$2.5	$10.0
56	Qwen 2.5 72B Qwen	47.9	28.0%	80.0%	1230	80	$0.36	$0.40
57	Grok 2 xAI	47.6	28.0%	78.0%	1250	80	$2.0	$10.0
58	GPT-4.5 OpenAI	47.2	38.0%	88.0%	1310	60	$75.0	$150.0
59	Claude 3.5 Haiku Anthropic	45.8	22.0%	74.0%	1180	170	$0.80	$4.0
60	Mistral Small 3.2 Mistral	43.7	20.0%	70.0%	1190	160	$0.07	$0.20
61	Llama 3.3 70B Meta	42.6	22.0%	72.0%	1210	90	$0.10	$0.32
62	GPT-4.1 Nano OpenAI	40.3	18.0%	65.0%	1120	200	$0.10	$0.40
63	Claude 3 Opus Anthropic	40.1	22.0%	78.0%	1240	25	$15.0	$75.0
64	GPT-OSS 120B OpenAI	40	—%	—%	—	339	$0.04	$0.17
65	GLM-5 Zhipu AI	39.9	—%	—%	1456	55	$0.60	$1.9
66	Command R+ Cohere	39.7	20.0%	72.0%	1200	60	$2.5	$10.0
67	MiniMax M2.5 MiniMax	39.6	—%	—%	1443	59	$0.27	$1.1
68	Nova Pro Amazon	38.9	18.0%	68.0%	1180	100	$0.80	$3.2
69	GPT-OSS 20B OpenAI	38.7	—%	—%	—	312	$0.03	$0.13
70	Gemma 3 27B Google	35.6	—%	—%	1339	58	$0.08	$0.45
71	Command R Cohere	32.3	12.0%	62.0%	1130	90	$0.15	$0.60
72	Ministral 3 8B Mistral	32.2	—%	—%	—	172	$0.15	$0.15
73	Phi-4 Mini Microsoft	31.1	—%	—%	—	150	$0.00	$0.00
74	Nova Lite Amazon	31.1	10.0%	55.0%	1110	150	$0.06	$0.24
75	Reka Flash 3 Reka AI	30.5	—%	—%	—	136	$0.10	$0.20
76	Phi-4 Microsoft	28.5	—%	—%	—	93	$0.07	$0.14
77	Gemma 3 12B Google	28.3	—%	—%	—	90	$0.05	$0.15
78	GLM-4.6V Zhipu AI	27.3	—%	—%	—	68	$0.30	$0.90
79	Kimi K2.5 Moonshot AI	26.1	—%	—%	—	45	$0.45	$2.3
80	Sonar Pro Perplexity	25.9	—%	—%	—	50	$3.0	$15.0
81	Jamba 1.5 Mini AI21 Labs	25.7	—%	—%	—	35	$0.20	$0.40
82	Gemma 3 4B Google	25	—%	—%	—	—	$0.05	$0.10
83	Yi Lightning 01.AI	25	—%	—%	—	—	$0.14	$0.14
84	Phi-4 Reasoning Plus Microsoft	25	—%	—%	—	—	$0.13	$0.50
85	GPT-5.6 Luna OpenAI	25	—%	—%	—	—	$0.20	$1.2
86	GPT-5.4 nano OpenAI	25	—%	—%	—	—	$0.20	$1.3
87	Gemini Embedding 001 Google	25	—%	—%	—	—	$0.15	$0.00
88	text-embedding-3-small OpenAI	25	—%	—%	—	—	$0.02	$0.00
89	text-embedding-3-large OpenAI	25	—%	—%	—	—	$0.13	$0.00
90	Lyria 3 Clip Preview Google	25	—%	—%	—	—	$0.00	$0.00
91	Qwen3-ASR Flash Alibaba	25	—%	—%	—	—	$0.04	$0.04
92	Qwen3.6 35B-A3B Alibaba	25	—%	—%	—	—	$0.10	$0.90
93	Qwen Plus Alibaba	25	—%	—%	—	—	$0.26	$0.78
94	Qwen3.6 Flash Alibaba	25	—%	—%	—	—	$0.19	$1.1
95	Qwen3.8 Flash Alibaba	25	—%	—%	—	—	$0.15	$0.47
96	Qwen3 8B Alibaba	25	—%	—%	—	—	$0.12	$0.46
97	Gemini Embedding 2 Google	25	—%	—%	—	—	$0.20	$0.00
98	Lyria 3 Pro Preview Google	25	—%	—%	—	—	$0.00	$0.00
99	DeepSeek V4 Flash Vision Exp DeepSeek	25	—%	—%	—	—	$0.22	$0.66
100	DeepSeek V4 Flash DeepSeek	25	—%	—%	—	—	$0.04	$0.11
101	DeepSeek V4.1 Flash DeepSeek	25	—%	—%	—	—	$0.30	$1.2
102	Muse Spark 1.2 Contributor Meta	25	—%	—%	—	—	$0.10	$0.20
103	Muse Spark 1.3 Contributor Meta	25	—%	—%	—	—	$0.10	$0.20
104	Pixtral 12B Mistral	25	—%	—%	—	—	$0.15	$0.15
105	Devstral Small Mistral	25	—%	—%	—	—	$0.10	$0.30
106	Mistral Embed Mistral	25	—%	—%	—	—	$0.10	$0.00
107	Devstral Small 2505 Mistral	25	—%	—%	—	—	$0.10	$0.30
108	Devstral Small 2 Mistral	25	—%	—%	—	—	$0.00	$0.00
109	Mistral 7B Mistral	25	—%	—%	—	—	$0.25	$0.25
110	Voxtral Small (latest) Mistral	25	—%	—%	—	—	$0.10	$0.30
111	Ministral 8B (latest) Mistral	25	—%	—%	—	—	$0.10	$0.10
112	Mistral Nemo Mistral	25	—%	—%	—	—	$0.02	$0.03
113	Mistral Small (latest) Mistral	25	—%	—%	—	—	$0.15	$0.60
114	Open Mistral Nemo Mistral	25	—%	—%	—	—	$0.15	$0.15
115	Mistral Small 4 Mistral	25	—%	—%	—	—	$0.15	$0.60
116	Ministral 3B (latest) Mistral	25	—%	—%	—	—	$0.04	$0.04
117	Qwen2.5 72B Instruct Alibaba	25	—%	—%	—	—	$0.36	$0.40
118	DeepSeek V4 Flash 0731 Alibaba	25	—%	—%	—	—	$0.06	$0.12
119	Qwen2.5-Omni 7B Alibaba	25	—%	—%	—	—	$0.10	$0.40
120	Qwen-MT Turbo Alibaba	25	—%	—%	—	—	$0.16	$0.49
121	Qwen3-Next 80B-A3B Instruct Alibaba	25	—%	—%	—	—	$0.09	$1.1
122	Qwen3 Coder Flash Alibaba	25	—%	—%	—	—	$0.20	$0.97
123	Qwen-VL Plus Alibaba	25	—%	—%	—	—	$0.21	$0.63
124	Qwen3.5 27B Alibaba	25	—%	—%	—	—	$0.20	$1.6
125	Qwen Flash Alibaba	25	—%	—%	—	—	$0.05	$0.40
126	Qwen Turbo Alibaba	25	—%	—%	—	—	$0.05	$0.20
127	Qwen3-VL 30B-A3B Alibaba	25	—%	—%	—	—	$0.20	$0.80
128	Qwen3-VL Plus Alibaba	25	—%	—%	—	—	$0.20	$1.6
129	Qwen2.5 7B Instruct Alibaba	25	—%	—%	—	—	$0.10	$0.20
130	Qwen3-Coder 30B-A3B Instruct Alibaba	25	—%	—%	—	—	$0.07	$0.28
131	Gemma 2 2b It NVIDIA	25	—%	—%	—	—	$0.00	$0.00
132	Qwen3 14B Alibaba	25	—%	—%	—	—	$0.12	$0.24
133	GLM-5.3-Flash Zhipu AI	25	—%	—%	—	—	$0.15	$0.50
134	GLM-4.7-Flash Zhipu AI	25	—%	—%	—	—	$0.06	$0.40
135	GLM-4.7-FlashX Zhipu AI	25	—%	—%	—	—	$0.07	$0.40
136	GLM-4.5-Flash Zhipu AI	25	—%	—%	—	—	$0.00	$0.00
137	Qwen-Omni Turbo Alibaba	25	—%	—%	—	—	$0.07	$0.27
138	Command R7B Arabic Cohere	25	—%	—%	—	—	$0.04	$0.15
139	North Mini Code Cohere	25	—%	—%	—	—	$0.00	$0.00
140	Command R7B Cohere	25	—%	—%	—	—	$0.04	$0.15
141	MiniMax-M2 MiniMax	25	—%	—%	—	—	$0.26	$1.0
142	GLM-4.5-Air Zhipu AI	25	—%	—%	—	—	$0.13	$0.85
143	MiniMax-M3 MiniMax	25	—%	—%	—	—	$0.30	$1.2
144	MiniMax-M2.7 MiniMax	25	—%	—%	—	—	$0.30	$1.2
145	Step 3.5 Flash StepFun	25	—%	—%	—	—	$0.10	$0.30
146	Step 3.5 Flash 2603 StepFun	25	—%	—%	—	—	$0.10	$0.30
147	Qwen Image NVIDIA	25	—%	—%	—	—	$0.00	$0.00
148	Qwen Image Edit NVIDIA	25	—%	—%	—	—	$0.00	$0.00
149	Laguna XS 2.1 NVIDIA	25	—%	—%	—	—	$0.06	$0.12
150	Mistral Large 3 675B Instruct 2512 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
151	Ministral 3 14B Instruct 2512 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
152	mistral-small-4-119b-2603 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
153	Mistral: Mixtral 8x7B Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
154	Mistral-7B-Instruct-v0.3 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
155	Magistral Small 2506 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
156	streampetr NVIDIA	25	—%	—%	—	—	$0.00	$0.00
157	Llama 3.3 Nemotron Super 49B v1.5 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
158	usdcode NVIDIA	25	—%	—%	—	—	$0.00	$0.00
159	Nemotron 3 Nano Omni NVIDIA	25	—%	—%	—	—	$0.00	$0.00
160	cosmos-transfer1-7b NVIDIA	25	—%	—%	—	—	$0.00	$0.00
161	nemotron-voicechat NVIDIA	25	—%	—%	—	—	$0.00	$0.00
162	studiovoice NVIDIA	25	—%	—%	—	—	$0.00	$0.00
163	nemotron-3-content-safety NVIDIA	25	—%	—%	—	—	$0.00	$0.00
164	cosmos-transfer2.5-2b NVIDIA	25	—%	—%	—	—	$0.00	$0.00
165	bevformer NVIDIA	25	—%	—%	—	—	$0.00	$0.00
166	Llama 3.1 Nemotron Nano VL 8B v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
167	llama-nemotron-embed-vl-1b-v2 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
168	synthetic-video-detector NVIDIA	25	—%	—%	—	—	$0.00	$0.00
169	Nemotron 3 Super NVIDIA	25	—%	—%	—	—	$0.08	$0.40
170	llama-nemotron-rerank-vl-1b-v2 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
171	usdvalidate NVIDIA	25	—%	—%	—	—	$0.00	$0.00
172	Active Speaker Detection NVIDIA	25	—%	—%	—	—	$0.00	$0.00
173	Llama 3.1 Nemotron Ultra 253B NVIDIA	25	—%	—%	—	—	$0.00	$0.00
174	llama-3_2-nemoretriever-300m-embed-v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
175	nv-embedcode-7b-v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
176	llama-3.1-nemotron-safety-guard-8b-v3 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
177	nemotron-mini-4b-instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
178	cosmos-predict1-5b NVIDIA	25	—%	—%	—	—	$0.00	$0.00
179	nemotron-content-safety-reasoning-4b NVIDIA	25	—%	—%	—	—	$0.00	$0.00
180	riva-translate-4b-instruct-v1_1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
181	BGE M3 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
182	sparsedrive NVIDIA	25	—%	—%	—	—	$0.00	$0.00
183	gliner-pii NVIDIA	25	—%	—%	—	—	$0.00	$0.00
184	Llama 3.1 Nemotron Nano 8B v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
185	nv-embed-v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
186	rerank-qa-mistral-4b NVIDIA	25	—%	—%	—	—	$0.00	$0.00
187	Nemotron 3.5 Lightning 30B A3B NVIDIA	25	—%	—%	—	—	$0.00	$0.00
188	nemotron-3-nano-30b-a3b NVIDIA	25	—%	—%	—	—	$0.05	$0.20
189	paligemma NVIDIA	25	—%	—%	—	—	$0.00	$0.00
190	Ministral 3 3B Amazon Bedrock	25	—%	—%	—	—	$0.10	$0.10
191	nvidia-nemotron-nano-9b-v2 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
192	Cosmos Reason2 8B NVIDIA	25	—%	—%	—	—	$0.00	$0.00
193	Gemma 3 4B IT NVIDIA	25	—%	—%	—	—	$0.00	$0.00
194	Gemma 3n E2b It NVIDIA	25	—%	—%	—	—	$0.00	$0.00
195	Gemma 3n E4b It NVIDIA	25	—%	—%	—	—	$0.00	$0.00
196	Llama 3.1 8B Instruct NVIDIA	25	—%	—%	—	—	$0.05	$0.08
197	Llama Guard 4 12B NVIDIA	25	—%	—%	—	—	$0.18	$0.18
198	Llama-3.2-90B-Vision-Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
199	Llama 3.2 1b Instruct NVIDIA	25	—%	—%	—	—	$0.03	$0.20
200	esmfold NVIDIA	25	—%	—%	—	—	$0.00	$0.00
201	Llama 3.2 11b Vision Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
202	Llama 3.1 70b Instruct NVIDIA	25	—%	—%	—	—	$0.40	$0.40
203	Llama 3.3 70b Instruct NVIDIA	25	—%	—%	—	—	$0.10	$0.32
204	ByteDance-Seed/Seed-OSS-36B-Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
205	sarvam-m NVIDIA	25	—%	—%	—	—	$0.00	$0.00
206	Phi 4 Multimodal NVIDIA	25	—%	—%	—	—	$0.00	$0.00
207	Llama 3.1 Nemotron 70B Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
208	dracarys-llama-3.1-70b-instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
209	Whisper Large v3 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
210	solar-10.7b-instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
211	FLUX.1-schnell NVIDIA	25	—%	—%	—	—	$0.00	$0.00
212	Gemma 3 12B IT NVIDIA	25	—%	—%	—	—	$0.00	$0.00
213	Llama 3.2 3B Instruct NVIDIA	25	—%	—%	—	—	$0.05	$0.33
214	Llama 4 Maverick 17b 128e Instruct NVIDIA	25	—%	—%	—	—	$0.00	$0.00
215	esm2-650m NVIDIA	25	—%	—%	—	—	$0.00	$0.00
216	FLUX.2 Klein 4B NVIDIA	25	—%	—%	—	—	$0.00	$0.00
217	FLUX.1-dev NVIDIA	25	—%	—%	—	—	$0.00	$0.00
218	FLUX.1-Kontext-dev NVIDIA	25	—%	—%	—	—	$0.00	$0.00
219	Mercury 2.5 Inception	25	—%	—%	—	—	$0.04	$0.15
220	Mercury Edit 2 Inception	25	—%	—%	—	—	$0.25	$0.75
221	Mercury 2 Inception	25	—%	—%	—	—	$0.25	$0.75
222	solar-mini Upstage	25	—%	—%	—	—	$0.15	$0.15
223	Solar Pro 4 Upstage	25	—%	—%	—	—	$0.09	$0.36
224	solar-pro3 Upstage	25	—%	—%	—	—	$0.15	$0.60
225	solar-pro2 Upstage	25	—%	—%	—	—	$0.25	$0.25
226	Qwen3 235B A22B Instruct Vertex	25	—%	—%	—	—	$0.22	$0.88
227	Qwen3 Coder Next Amazon Bedrock	25	—%	—%	—	—	$0.12	$0.80
228	Qwen3 235B-A22B Instruct 2507 Amazon Bedrock	25	—%	—%	—	—	$0.09	$0.35
229	Voxtral Mini 3B 2507 Amazon Bedrock	25	—%	—%	—	—	$0.04	$0.04
230	NVIDIA Nemotron Nano 12B v2 VL BF16 Amazon Bedrock	25	—%	—%	—	—	$0.20	$0.60
231	NVIDIA Nemotron Nano 3 30B Amazon Bedrock	25	—%	—%	—	—	$0.06	$0.24
232	Gemma 4 E2B IT Amazon Bedrock	25	—%	—%	—	—	$0.04	$0.08
233	Llama 4 Maverick 17B Instruct (US) Amazon Bedrock	25	—%	—%	—	—	$0.24	$0.97
234	Ministral 14B 3.0 Amazon Bedrock	25	—%	—%	—	—	$0.20	$0.20
235	GPT OSS Safeguard 20B Amazon Bedrock	25	—%	—%	—	—	$0.07	$0.30
236	Llama 4 Scout 17B Instruct (US) Amazon Bedrock	25	—%	—%	—	—	$0.17	$0.66
237	NVIDIA Nemotron 3 Super 120B A12B Amazon Bedrock	25	—%	—%	—	—	$0.15	$0.65
238	GPT OSS Safeguard 120B Amazon Bedrock	25	—%	—%	—	—	$0.15	$0.60
239	Voxtral Small 24B 2507 Amazon Bedrock	25	—%	—%	—	—	$0.10	$0.30
240	Mistral Small 3.1 Azure	25	—%	—%	—	—	$0.10	$0.30
241	Phi-4-mini-reasoning Azure	25	—%	—%	—	—	$0.07	$0.30
242	Llama 4 Maverick 17B 128E Instruct FP8 Azure	25	—%	—%	—	—	$0.25	$1.0
243	Embed v3 English Azure	25	—%	—%	—	—	$0.10	$0.00
244	Embed v4 Azure	25	—%	—%	—	—	$0.12	$0.00
245	Gemma 3 27B IT Amazon Bedrock	25	—%	—%	—	—	$0.23	$0.38
246	Llama 4 Scout 17B 16E Instruct Azure	25	—%	—%	—	—	$0.20	$0.78
247	Model Router Azure	25	—%	—%	—	—	$0.14	$0.00
248	Embed v3 Multilingual Azure	25	—%	—%	—	—	$0.10	$0.00
249	Phi-4-reasoning Azure	25	—%	—%	—	—	$0.13	$0.50
250	Codestral 25.01 Azure	25	—%	—%	—	—	$0.30	$0.90
251	text-embedding-ada-002 OpenAI	25	—%	—%	—	—	$0.10	$0.00
252	Qwen-Omni Turbo Realtime Alibaba	25	—%	—%	—	—	$0.27	$1.1
253	MiniMax-M2.1 MiniMax	25	—%	—%	—	—	$0.30	$1.2
254	Step 3.7 Flash StepFun	25	—%	—%	—	—	$0.20	$1.1
255	mistral-nemotron NVIDIA	25	—%	—%	—	—	$0.00	$0.00
256	magpie-tts-zeroshot NVIDIA	25	—%	—%	—	—	$0.00	$0.00
257	Nemotron Nano 12B v2 VL NVIDIA	25	—%	—%	—	—	$0.00	$0.00
258	Llama 3.3 Nemotron Super 49B v1 NVIDIA	25	—%	—%	—	—	$0.00	$0.00
259	Nova Micro (US) Amazon Bedrock	25	—%	—%	—	—	$0.04	$0.14
260	Sonar Perplexity	24.9	—%	—%	—	—	$1.0	$1.0
261	Gemini 3.6 Flash Google	24.9	—%	—%	—	—	$0.75	$3.8
262	Gemini 3.1 Flash Lite Google	24.9	—%	—%	—	—	$0.25	$1.5
263	Nano Banana 2 Google	24.9	—%	—%	—	—	$0.50	$3.0
264	GPT-3.5-turbo OpenAI	24.9	—%	—%	—	—	$1.0	$2.0
265	Nano Banana 2 Lite Google	24.9	—%	—%	—	—	$0.25	$1.5
266	Qwen3-Omni Flash Alibaba	24.9	—%	—%	—	—	$0.43	$1.7
267	Qwen2.5-VL 72B Instruct Alibaba	24.9	—%	—%	—	—	$0.80	$1.0
268	Qwen3.5 122B-A10B Alibaba	24.9	—%	—%	—	—	$0.26	$2.1
269	Qwen3-Omni Flash Realtime Alibaba	24.9	—%	—%	—	—	$0.52	$2.0
270	Qwen2.5-VL 7B Instruct Alibaba	24.9	—%	—%	—	—	$0.35	$1.1
271	Qwen2.5 14B Instruct Alibaba	24.9	—%	—%	—	—	$0.35	$1.4
272	Qwen Plus Character (Japanese) Alibaba	24.9	—%	—%	—	—	$0.50	$1.4
273	Qwen3 235B-A22B Alibaba	24.9	—%	—%	—	—	$0.46	$1.8
274	Qwen3.6 27B Alibaba	24.9	—%	—%	—	—	$0.30	$2.0
275	Gemini Flash-Lite Latest Google	24.9	—%	—%	—	—	$0.30	$2.5
276	Gemini 3.8 Flash Google	24.9	—%	—%	—	—	$0.75	$3.8
277	Gemini 3.7 Flash Google	24.9	—%	—%	—	—	$0.75	$3.8
278	Gemini Flash Latest Google	24.9	—%	—%	—	—	$0.75	$3.8
279	DeepSeek V4 Pro DeepSeek	24.9	—%	—%	—	—	$0.43	$0.87
280	Magistral Small Mistral	24.9	—%	—%	—	—	$0.50	$1.5
281	Devstral 2 Mistral	24.9	—%	—%	—	—	$0.40	$2.0
282	Mixtral 8x7B Mistral	24.9	—%	—%	—	—	$0.70	$0.70
283	GLM-5.2 Mistral	24.9	—%	—%	—	—	$0.68	$2.1
284	Mistral Medium 3 Mistral	24.9	—%	—%	—	—	$0.40	$2.0
285	Mistral Large 3 Mistral	24.9	—%	—%	—	—	$0.50	$1.5
286	Grok Build 0.1 xAI	24.9	—%	—%	—	—	$1.0	$2.0
287	Qwen3 Coder Plus Alibaba	24.9	—%	—%	—	—	$0.65	$3.3
288	Qwen-VL Max Alibaba	24.9	—%	—%	—	—	$0.80	$3.2
289	Qwen3.6 Plus Alibaba	24.9	—%	—%	—	—	$0.33	$1.9
290	Qwen3.5 35B-A3B Alibaba	24.9	—%	—%	—	—	$0.31	$1.3
291	Qwen-VL OCR Alibaba	24.9	—%	—%	—	—	$0.72	$0.72
292	QwQ Plus Alibaba	24.9	—%	—%	—	—	$0.80	$2.4
293	Qwen3.5 397B-A17B Alibaba	24.9	—%	—%	—	—	$0.55	$3.5
294	Qwen2.5 32B Instruct Alibaba	24.9	—%	—%	—	—	$0.70	$2.8
295	Qwen3-VL 235B-A22B Alibaba	24.9	—%	—%	—	—	$0.70	$2.8
296	Qwen3.5 Plus Alibaba	24.9	—%	—%	—	—	$0.40	$2.4
297	GLM-5.1 Zhipu AI	24.9	—%	—%	—	—	$0.97	$3.0
298	GLM-4.5V Zhipu AI	24.9	—%	—%	—	—	$0.60	$1.8
299	GLM-4.5 Zhipu AI	24.9	—%	—%	—	—	$0.60	$2.2
300	GLM-4.6 Zhipu AI	24.9	—%	—%	—	—	$0.43	$1.8
301	GLM-4.7 Zhipu AI	24.9	—%	—%	—	—	$0.40	$1.8
302	Kimi K2.7 Code Moonshot AI	24.9	—%	—%	—	—	$0.71	$3.5
303	MiniMax-M2.7-highspeed MiniMax	24.9	—%	—%	—	—	$0.60	$2.4
304	Qwen2.5 Coder 32b Instruct NVIDIA	24.9	—%	—%	—	—	$0.66	$1.0
305	DeepSeek V4 Pro 0813 NVIDIA	24.9	—%	—%	—	—	$0.98	$3.0
306	MiniMax-M2.5-highspeed MiniMax	24.9	—%	—%	—	—	$0.60	$2.4
307	Nemotron 3 Ultra 550B A55B NVIDIA	24.9	—%	—%	—	—	$0.50	$2.5
308	Kimi K2 Thinking Vertex	24.9	—%	—%	—	—	$0.60	$2.5
309	Morph v3 Large Morph	24.9	—%	—%	—	—	$0.90	$1.9
310	Morph v3 Fast Morph	24.9	—%	—%	—	—	$0.80	$1.2
311	Auto Morph	24.9	—%	—%	—	—	$0.85	$1.6
312	Kimi K2 0905 NVIDIA	24.9	—%	—%	—	—	$0.60	$2.5
313	Nova 2 Lite (EU) Amazon Bedrock	24.9	—%	—%	—	—	$0.30	$2.5
314	Qwen3 VL 235B A22B Instruct Amazon Bedrock	24.9	—%	—%	—	—	$0.21	$1.9
315	Magistral Small 1.2 Amazon Bedrock	24.9	—%	—%	—	—	$0.50	$1.5
316	Devstral 2 123B Amazon Bedrock	24.9	—%	—%	—	—	$0.40	$2.0
317	GPT-3.5 Turbo 1106 Azure	24.9	—%	—%	—	—	$1.0	$2.0
318	GPT-3.5 Turbo 0125 Azure	24.9	—%	—%	—	—	$0.50	$1.5
319	DeepSeek-V3.2-Speciale Azure	24.9	—%	—%	—	—	$0.58	$1.7
320	Nano Banana Google	24.9	—%	—%	—	—	$0.30	$2.5
321	Gemini 3.5 Flash Lite Google	24.9	—%	—%	—	—	$0.30	$2.5
322	Devstral Medium Mistral	24.9	—%	—%	—	—	$0.40	$2.0
323	Qwen3.7 Plus Alibaba	24.9	—%	—%	—	—	$0.32	$1.3
324	Muse Glimmer 30B NVIDIA	24.9	—%	—%	—	—	$0.35	$1.5
325	GPT-5.4 mini OpenAI	24.8	—%	—%	—	—	$0.75	$4.5
326	Gemini 3.1 Flash Live Preview Google	24.8	—%	—%	—	—	$0.75	$4.5
327	Qwen3.6 Max Preview Alibaba	24.8	—%	—%	—	—	$1.0	$6.2
328	Muse Spark 1.1 Meta	24.8	—%	—%	—	—	$1.3	$4.3
329	Muse Spark 1.2 Meta	24.8	—%	—%	—	—	$1.3	$4.3
330	Grok 4.3 xAI	24.8	—%	—%	—	—	$1.3	$2.5
331	Grok 4.20 (Reasoning) xAI	24.8	—%	—%	—	—	$1.3	$2.5
332	Grok 4.20 Multi-Agent xAI	24.8	—%	—%	—	—	$1.3	$2.5
333	Qwen3.7 Max Alibaba	24.8	—%	—%	—	—	$1.5	$4.4
334	Qwen3-Next 80B-A3B (Thinking) Alibaba	24.8	—%	—%	—	—	$0.50	$6.0
335	QVQ Max Alibaba	24.8	—%	—%	—	—	$1.2	$4.8
336	GLM-5.3 Zhipu AI	24.8	—%	—%	—	—	$1.4	$4.4
337	GLM-5V-Turbo Zhipu AI	24.8	—%	—%	—	—	$1.2	$4.0
338	Kimi K2.6 Moonshot AI	24.8	—%	—%	—	—	$0.95	$4.0
339	Inkling NVIDIA	24.8	—%	—%	—	—	$1.0	$4.0
340	Sonar Reasoning Pro Perplexity	24.8	—%	—%	—	22	$2.0	$8.0
341	Gemini 2.5 Flash TTS Vertex	24.8	—%	—%	—	—	$0.50	$10.0
342	Palmyra X5 (US) Amazon Bedrock	24.8	—%	—%	—	—	$0.60	$6.0
343	Codex Mini Azure	24.8	—%	—%	—	—	$1.5	$6.0
344	GPT-3.5 Turbo Instruct Azure	24.8	—%	—%	—	—	$1.5	$2.0
345	Muse Spark 1.3 Meta	24.8	—%	—%	—	—	$1.3	$4.3
346	Jamba 1.5 Large AI21 Labs	24.7	—%	—%	—	19	$2.0	$8.0
347	Gemini 2.5 Computer Use Preview 10-2025 Google	24.7	—%	—%	—	—	$1.3	$10.0
348	Gemini 3.5 Flash Google	24.7	—%	—%	—	—	$1.5	$9.0
349	Qwen3.8 Max Alibaba	24.7	—%	—%	—	—	$2.0	$6.0
350	Magistral Medium (latest) Mistral	24.7	—%	—%	—	—	$2.0	$5.0
351	Mixtral 8x22B Mistral	24.7	—%	—%	—	—	$2.0	$6.0
352	Mistral Medium (latest) Mistral	24.7	—%	—%	—	—	$1.5	$7.5
353	Mistral Large (latest) Mistral	24.7	—%	—%	—	—	$2.0	$6.0
354	Mistral Large 2.1 Mistral	24.7	—%	—%	—	—	$2.0	$6.0
355	Grok 4.5 xAI	24.7	—%	—%	—	—	$2.0	$6.0
356	Grok 4.6 xAI	24.7	—%	—%	—	—	$2.0	$6.0
357	Qwen Max Alibaba	24.7	—%	—%	—	—	$1.6	$6.4
358	Qwen3-Coder 480B-A35B Instruct Alibaba	24.7	—%	—%	—	—	$1.5	$7.5
359	Mistral: Mixtral 8x22B Instruct NVIDIA	24.7	—%	—%	—	—	$2.0	$6.0
360	Perplexity Sonar Deep Research Perplexity	24.7	—%	—%	—	—	$2.0	$8.0
361	GPT-5-Codex Azure	24.7	—%	—%	—	—	$1.3	$10.0
362	GPT-5.1 Codex Max Azure	24.7	—%	—%	—	—	$1.3	$10.0
363	Mistral Medium 3.5 Mistral	24.7	—%	—%	—	—	$1.5	$7.5
364	Kimi K2.7 Code HighSpeed Moonshot AI	24.7	—%	—%	—	—	$1.9	$8.0
365	Claude Sonnet 5 Anthropic	24.6	—%	—%	—	—	$2.0	$10.0
366	GPT-5.6 Sol OpenAI	24.6	—%	—%	—	—	$2.0	$10.0
367	GPT-5.3 Codex OpenAI	24.6	—%	—%	—	—	$1.8	$14.0
368	Deep Research Max Preview (Apr-21-2026) Google	24.6	—%	—%	—	—	$2.0	$12.0
369	Nano Banana Pro Google	24.6	—%	—%	—	—	$2.0	$12.0
370	Deep Research Preview (Apr-21-2026) Google	24.6	—%	—%	—	—	$2.0	$12.0
371	GPT-5.6 Terra OpenAI	24.6	—%	—%	—	—	$2.0	$12.0
372	GPT-5.2 Chat OpenAI	24.6	—%	—%	—	—	$1.8	$14.0
373	GPT-5.3 Chat (latest) OpenAI	24.6	—%	—%	—	—	$1.8	$14.0
374	Qwen-MT Plus Alibaba	24.6	—%	—%	—	—	$2.5	$7.4
375	Command A Plus Cohere	24.6	—%	—%	—	—	$2.5	$10.0
376	Command A Reasoning Cohere	24.6	—%	—%	—	—	$2.5	$10.0
377	Command A Vision Cohere	24.6	—%	—%	—	—	$2.5	$10.0
378	Command A Translate Cohere	24.6	—%	—%	—	—	$2.5	$10.0
379	Step 1 (32K) StepFun	24.6	—%	—%	—	—	$2.0	$9.6
380	Palmyra X4 (US) Amazon Bedrock	24.6	—%	—%	—	—	$2.5	$10.0
381	GPT-5.2 Codex Azure	24.6	—%	—%	—	—	$1.8	$14.0
382	GPT-5.3 Codex Spark OpenAI	24.6	—%	—%	—	—	$1.8	$14.0
383	GPT-5.4 OpenAI	24.5	—%	—%	—	—	$2.5	$15.0
384	Gemini 3.1 Flash TTS Preview Google	24.5	—%	—%	—	—	$1.0	$20.0
385	Gemini Omni Flash Preview Google	24.5	—%	—%	—	—	$1.5	$17.5
386	Kimi K3 Moonshot AI	24.5	—%	—%	—	—	$2.6	$13.3
387	Gemini 2.5 Pro TTS Vertex	24.5	—%	—%	—	—	$1.0	$20.0
388	Nova Premier (US) Amazon Bedrock	24.5	—%	—%	—	—	$2.5	$12.5
389	AU Anthropic Claude Sonnet 4.6 Amazon Bedrock	24.4	—%	—%	—	—	$3.3	$16.5
390	GPT-5.6 OpenAI	24.3	—%	—%	—	—	$4.0	$20.0
391	Gemini 3.5 Live Translate Preview Google	24.3	—%	—%	—	—	$3.5	$21.0
392	GPT-Realtime-2.1 OpenAI	24.2	—%	—%	—	—	$4.0	$24.0
393	Step 2 (16K) StepFun	24.2	—%	—%	—	—	$5.2	$16.4
394	Claude Opus 4.7 Anthropic	24.1	—%	—%	—	—	$5.0	$25.0
395	Claude Opus 4.8 Anthropic	24.1	—%	—%	—	—	$5.0	$25.0
396	Claude Opus 5 Anthropic	24.1	—%	—%	—	—	$5.0	$25.0
397	GPT-5.5 OpenAI	24	—%	—%	—	—	$5.0	$30.0
398	Qwen3-LiveTranslate Flash Realtime Alibaba	24	—%	—%	—	—	$10.0	$10.0
399	GPT Chat Latest Azure	24	—%	—%	—	—	$5.0	$30.0
400	gpt-image-2 OpenAI	24	—%	—%	—	—	$5.0	$30.0
401	GPT-4 Turbo OpenAI	23.5	—%	—%	—	—	$10.0	$30.0
402	GPT-4 Turbo Vision Azure	23.5	—%	—%	—	—	$10.0	$30.0
403	Claude Fable 5.1 Anthropic	23.1	—%	—%	—	—	$10.0	$50.0
404	GPT-6 Astra OpenAI	23.1	—%	—%	—	—	$10.0	$50.0
405	Claude Mythos 5 Azure	23.1	—%	—%	—	—	$10.0	$50.0
406	Claude Opus 4.1 Vertex	22.2	—%	—%	—	—	$15.0	$75.0
407	AU Anthropic Claude Opus 4.6 Amazon Bedrock	21.9	—%	—%	—	—	$16.5	$82.5
408	GPT-5 Pro OpenAI	21.3	—%	—%	—	—	$15.0	$120.0
409	GPT-4 OpenAI	21.3	—%	—%	—	—	$30.0	$60.0
410	GPT-5.2 Pro OpenAI	19.8	—%	—%	—	—	$21.0	$168.0
411	GPT-5.5 Pro OpenAI	18.8	—%	—%	—	—	$30.0	$180.0
412	GPT-5.4 Pro OpenAI	18.8	—%	—%	—	—	$30.0	$180.0
413	Doubao Seed 2.0 ByteDance	14	—%	—%	1474	—	—	—
414	Veo 3.1 Google	11.2	—%	—%	1401	—	—	—
415	Sora 2 OpenAI	9.9	—%	—%	1368	—	—	—
416	Nova 2.0 Lite Amazon	9.5	—%	—%	—	221	—	—
417	Veo 3 Google	8.8	—%	—%	1340	—	—	—
418	Seedream 4.5 ByteDance	7.9	—%	—%	1316	—	—	—
419	MiMo V2 Flash Xiaomi	6.4	—%	—%	—	155	—	—
420	Qwen 3 Next 80B Qwen	5.6	—%	—%	—	138	—	—
421	Ministral 3 14B Mistral	3	—%	—%	—	83	—	—
422	Ring Flash 2.0 InclusionAI	3	—%	—%	—	83	—	—
423	Nemotron 3 Nano NVIDIA	2.7	—%	—%	—	76	—	—
424	Step 2.5 Flash StepFun	2.3	—%	—%	—	67	—	—
425	K-EXAONE LG AI Research	2	—%	—%	—	62	—	—
426	Qwen 3 Coder 480B Qwen	1.9	—%	—%	—	60	—	—
427	Qwen 3 VL 235B Qwen	1.3	—%	—%	—	46	—	—
428	Magistral Medium 1.2 Mistral	0.5	—%	—%	—	29	—	—
429	DALL-E 3 OpenAI	0	—%	—%	—	—	—	—
430	Kling 2.5 Turbo Kuaishou	0	—%	—%	—	—	—	—
431	ERNIE 4.5 Baidu	0	—%	—%	—	—	—	—
432	ERNIE X1 Baidu	0	—%	—%	—	—	—	—
433	Hermes 4 70B Nous Research	0	—%	—%	—	—	—	—
434	Imagen 4 Google	0	—%	—%	—	—	—	—
435	Luma Ray 3 Luma AI	0	—%	—%	—	—	—	—
436	Runway Gen-4 Runway	0	—%	—%	—	—	—	—
437	Midjourney v7 Midjourney	0	—%	—%	—	—	—	—
438	Midjourney v6.1 Midjourney	0	—%	—%	—	—	—	—
439	Stable Diffusion 3.5 Stability AI	0	—%	—%	—	—	—	—
440	Flux 1.1 Pro Black Forest Labs	0	—%	—%	—	—	—	—
441	Flux 1.0 Dev Black Forest Labs	0	—%	—%	—	—	—	—
442	Ideogram 3.0 Ideogram	0	—%	—%	—	—	—	—
443	Pika 2.5 Pika	0	—%	—%	—	—	—	—
444	o1-pro OpenAI	0	—%	—%	—	—	$150.0	$600.0
445	chatgpt-image-latest OpenAI	0	—%	—%	—	—	—	—
446	gpt-image-1 OpenAI	0	—%	—%	—	—	—	—
447	gpt-image-1-mini OpenAI	0	—%	—%	—	—	—	—
448	gpt-image-1.5 OpenAI	0	—%	—%	—	—	—	—
449	Veo 3.1 fast Google	0	—%	—%	—	—	—	—
450	Gemma 4 26B A4B IT Google	0	—%	—%	—	—	—	—
451	Veo 3.1 lite Google	0	—%	—%	—	—	—	—
452	Gemma 4 31B IT Google	0	—%	—%	—	—	—	—
453	Voxtral Mini TTS (latest) Mistral	0	—%	—%	—	—	—	—
454	Voxtral Mini (latest) Mistral	0	—%	—%	—	—	—	—
455	Grok Imagine Video 1.5 xAI	0	—%	—%	—	—	—	—
456	Grok Imagine Image 2.0 xAI	0	—%	—%	—	—	—	—
457	Grok Imagine Image Quality xAI	0	—%	—%	—	—	—	—
458	Grok Imagine Video xAI	0	—%	—%	—	—	—	—
459	Aya Expanse 32B Cohere	0	—%	—%	—	—	—	—
460	Aya Expanse 8B Cohere	0	—%	—%	—	—	—	—
461	Aya Vision 8B Cohere	0	—%	—%	—	—	—	—
462	Aya Vision 32B Cohere	0	—%	—%	—	—	—	—
463	Step TTS 2 StepFun	0	—%	—%	—	—	—	—
464	StepAudio 2.5 ASR StepFun	0	—%	—%	—	—	—	—
465	StepAudio 2.5 TTS StepFun	0	—%	—%	—	—	—	—
466	Grok Imagine Image xAI	0	—%	—%	—	—	—	—

Data sourced from Chatbot Arena, OpenRouter, and public benchmarks. Updated daily. Scores are dynamically computed based on your selected metrics.

How It Works

Choose a Persona or Build Your Own

Select a preset (Developer, Researcher, Business, etc.) that pre-selects relevant benchmarks and weights. Or customize everything from scratch.

See Real Data, Not Abstract Scores

Every metric shows the actual value — ELO 1410, $2.50/1M tokens, 65 tok/s. No normalization black box. The raw numbers are always visible.

Dynamic Ranking — Your Weights, Your Score

For each metric you select, we find the min/max across all models, normalize to 0-100, then compute a weighted composite. Missing data is excluded and weights renormalize automatically.

Share Your Rankings

Your exact configuration is encoded in the URL. Share it with your team or embed it — they'll see your exact ranking.

Data sourced from Chatbot Arena, OpenRouter, SWE-bench, and public research papers.

45 benchmarks across General, Coding, Math, Reasoning, Speed, Cost, and Context categories.

Frequently Asked Questions

What is the best AI model in 2026?

The best AI model depends on your use case. As of 2026, top contenders include Gemini 3.1 Pro, Claude Opus 4.6, and GPT-5.2 for general intelligence. For coding, Claude Opus 4.6 and GPT-5.2 lead on SWE-bench. For budget-conscious users, DeepSeek V3.2 and Gemini 2.5 Flash offer excellent performance per dollar. Use our rankings to sort models by YOUR priorities.

How do AI benchmarks work?

AI benchmarks are standardized tests that evaluate language models across specific capabilities. Common benchmarks include Chatbot Arena ELO (human preference voting), SWE-bench (real software engineering tasks), MMLU-Pro (knowledge and reasoning), GPQA Diamond (graduate-level science), and MATH (competition mathematics). Each benchmark tests a different aspect of model capability, and no single benchmark tells the whole story.

What is Chatbot Arena ELO?

Chatbot Arena ELO is a human preference ranking system where users compare AI model responses in blind head-to-head matchups. The ELO rating (borrowed from chess) reflects how often a model is preferred over others. Higher ELO means the model is more frequently preferred. It's considered one of the most reliable benchmarks because it uses real human judgment rather than automated scoring.

Which AI model is best for coding?

For coding tasks in 2026, the top models are Claude Opus 4.6 (80.8% SWE-bench), GPT-5.2 (80.0% SWE-bench), and Gemini 3.1 Pro (80.6% SWE-bench). For more affordable coding, GPT-5.1 Codex and Qwen 3 Coder offer strong performance at lower costs. The Developer persona pre-weights coding benchmarks to help you find the best fit.

Which AI model is cheapest?

The cheapest AI models by API pricing include GPT-5 Nano ($0.05/1M input), GPT-4.1 Nano ($0.10/1M input), Nova Lite ($0.06/1M input), and Gemini 2.5 Flash Lite ($0.10/1M input). For the best balance of quality and cost, DeepSeek V3.2 ($0.14/1M input) and Qwen 3 32B ($0.10/1M input) offer strong performance at budget prices.