What is the cheapest AI model for chat?

The cheapest AI chat models are typically smaller models like GPT-3.5-turbo, Claude Instant, or open-source models on platforms like Together AI and Groq. Prices can be as low as $0.10-0.25 per million tokens. However, cheaper models may have reduced capabilities compared to flagship models like GPT-4 or Claude 3 Opus. Use CloudPrice to compare current prices across all providers.

How much does GPT-4 cost?

GPT-4 pricing varies by version. GPT-4 Turbo costs approximately $10 per million input tokens and $30 per million output tokens. GPT-4o is more affordable at around $5/$15 per million tokens. OpenAI also offers GPT-4o-mini at significantly lower prices. Pricing may change, so check CloudPrice for current rates.

What is the difference between input and output tokens?

Input tokens are the text you send to the AI model (your prompts, context, and conversation history). Output tokens are what the model generates in response. Most AI providers charge differently for input and output tokens, with output tokens typically being 2-4x more expensive because they require more computation to generate.

Which AI model has the largest context window?

Google's Gemini 1.5 Pro currently offers one of the largest context windows at 1 million tokens. Anthropic's Claude 3 models support 200K tokens, while OpenAI's GPT-4 Turbo supports 128K tokens. Larger context windows allow processing more text in a single request but may cost more.

What does 'supports vision' mean for AI models?

AI models with vision support can process and understand images in addition to text. You can send images as part of your prompt and the model can describe, analyze, or answer questions about them. Examples include GPT-4 Vision, Claude 3, and Gemini Pro Vision. Vision capabilities typically cost extra per image processed.

AI Models Pricing & Benchmarks Comparison - 1900+ LLM Models

Compare AI model pricing and benchmarks across providers including OpenAI, Anthropic, Google, AWS Bedrock, Azure, Mistral, and more. Filter by model capabilities like vision, function calling, and reasoning support. Find the most cost-effective model for your use case. Currently tracking 1,909 models across 103 providers.

The data is based on LiteLLM, maintained by the open-source community, and benchmark data from Artificial Analysis. The latest update occurred on April 15, 2026 at 12:00 AM UTC

Model	Provider	Input Price, $	Output Price, $	Price Compare	Context	Max Output	Intelligence	Coding
Zai.glm 5	AWS Bedrock	1.00	3.20	compare	200K	128K	49.8#5	44.2#11
Moonshotai.kimi K2.5	AWS Bedrock	0.600	3.00	compare	262K	262K	46.8#11	39.5#18
Claude Opus 4.6	AWS Bedrock	5.00	25.00	compare	1.0M	128K	46.5#12	47.6#6
Claude Sonnet 4.6	AWS Bedrock	3.00	15.00	compare	1.0M	64K	44.4#16	46.4#9
Claude Opus 4.5	AWS Bedrock	5.00	25.00	compare	200K	64K	43.1#18	42.9#14
Zai.glm 4.7	AWS Bedrock	0.600	2.20	compare	200K	128K	42.1#20	36.3#28
Minimax.minimax M2.5	AWS Bedrock	0.360	1.44	compare	1.0M	8K	41.9#22	37.4#24
Claude Sonnet 4.5	AWS Bedrock	3.00	15.00	compare	200K	64K	37.1#32	33.5#35
Minimax.minimax M2	AWS Bedrock	0.300	1.20	compare	128K	8K	36.1#34	29.2#44
Claude Opus 4.1	AWS Bedrock	15.00	75.00	compare	200K	32K	36.0#35	N/A
GPT-oss-120b	AWS Bedrock	0.150	0.600	compare	131K	33K	33.3#39	28.6#45
Claude Sonnet 4.20250514	AWS Bedrock	3.00	15.00	compare	1.0M	64K	33.0#41	30.6#37
Claude Opus 4.20250514	AWS Bedrock	15.00	75.00	compare	200K	32K	33.0#41	N/A
V3.2	AWS Bedrock	0.740	2.22	compare	164K	164K	32.1#43	34.6#33
Claude Haiku 4.5	AWS Bedrock	1.00	5.00	compare	200K	64K	31.1#46	29.6#43
Eu.anthropic.claude 3 7 Sonnet 20250219 V1	AWS Bedrock	3.00	15.00	compare	200K	8K	30.8#48	26.7#47
Zai.glm 4.7 Flash	AWS Bedrock	0.070	0.400	compare	200K	128K	30.1#53	25.9#50
Qwen3 Coder Next	AWS Bedrock	0.600	1.44	compare	262K	8K	28.3#58	22.9#63
Us.deepseek.r1 V1	AWS Bedrock	1.35	5.40	compare	128K	4K	27.1#60	24.0#57
Qwen3 235B A22b 2507 V1	AWS Bedrock	0.220	0.880	compare	262K	131K	25.0#70	22.1#65
Qwen3 Coder 480B A35b V1	AWS Bedrock	0.220	1.80	compare	262K	66K	24.8#71	24.6#56
GPT-oss-20b	AWS Bedrock	0.075	0.300	compare	131K	33K	24.5#72	18.5#80
Moonshotai.kimi K2 Thinking	AWS Bedrock	0.730	3.03	compare	262K	262K	24.1#73	15.5#96
Moonshotai.kimi K2.5	AWS Bedrock	0.720	3.60	compare	262K	262K	23.1#77	12.6#116
Nova 2 Pro Preview 20251202 V1	AWS Bedrock	2.19	17.50	compare	1.0M	64K	23.1#77	20.5#71
Mistral Large 3 675B Instruct	AWS Bedrock	0.500	1.50	compare	128K	8K	22.8#81	22.7#64
Qwen3 Vl 235B A22b	AWS Bedrock	0.530	2.66	compare	128K	8K	20.8#87	16.5#89
Qwen3 Next 80B A3b	AWS Bedrock	0.150	1.20	compare	128K	8K	20.1#90	15.3#97
Qwen3 Coder 30B A3b V1	AWS Bedrock	0.150	0.600	compare	262K	131K	20.0#91	19.4#78
Us.amazon.nova Premier V1	AWS Bedrock	2.50	12.50	compare	1.0M	10K	19.0#95	13.8#108
Eu.anthropic.claude 3 5 Haiku 20241022 V1	AWS Bedrock	0.250	1.25	compare	200K	8K	18.7#98	10.7#131
Magistral Small 2509	AWS Bedrock	0.500	1.50	compare	128K	8K	18.2#102	14.8#98
Claude 3 Opus 20240229 V1	AWS Bedrock	15.00	75.00	compare	200K	4K	18.0#162	19.5#75
Nova 2 Lite V1	AWS Bedrock	0.300	2.50	compare	1.0M	64K	18.0#104	12.5#117
Llama3 1 405B Instruct V1	AWS Bedrock	5.32	16.00	compare	128K	4K	17.4#108	14.5#100
V3 V1	AWS Bedrock	0.580	1.68	compare	164K	82K	16.5#116	16.4#90
Ministral 3 14B Instruct	AWS Bedrock	0.200	0.200	compare	128K	8K	16.0#120	10.9#127
Claude 3 5 Sonnet 20240620 V1	AWS Bedrock	3.00	15.00	compare	1.0M	4K	15.9#123	30.2#40
Ministral 3 8B Instruct	AWS Bedrock	0.150	0.150	compare	128K	8K	14.8#134	10.0#136
Qwen3 32B V1	AWS Bedrock	0.150	0.600	compare	131K	16K	14.5#137	N/A
Llama3 3 70B Instruct V1	AWS Bedrock	0.720	0.720	compare	128K	4K	14.5#137	10.7#131
Nova Pro V1	AWS Bedrock	0.800	3.20	compare	300K	10K	13.5#145	11.0#126
Nvidia.nemotron Nano 9B V2	AWS Bedrock	0.060	0.230	compare	128K	8K	13.2#150	7.5#147
Nvidia.nemotron Nano 3 30B	AWS Bedrock	0.060	0.240	compare	262K	8K	13.2#150	15.8#93
Mistral Large 2407 V1	AWS Bedrock	3.00	9.00	compare	128K	8K	13.0#153	N/A
Nova Lite V1	AWS Bedrock	0.060	0.240	compare	300K	10K	12.7#158	5.1#156
Llama3 1 70B Instruct V1	AWS Bedrock	0.990	0.990	compare	128K	2K	12.5#162	10.9#127
Claude 3 Haiku 20240307 V1	AWS Bedrock	0.250	1.25	compare	200K	4K	12.3#167	6.7#152
Llama3 2 90B Instruct V1	AWS Bedrock	2.00	2.00	compare	128K	4K	11.9#174	N/A
Llama3 1 8B Instruct V1	AWS Bedrock	0.220	0.220	compare	128K	2K	11.8#175	4.9#157
Ministral 3 3B Instruct	AWS Bedrock	0.100	0.100	compare	128K	8K	11.2#176	4.8#158
Jamba 1 5 Large V1	AWS Bedrock	2.00	8.00	compare	256K	256K	10.7#179	N/A
Gemma 3 27B It	AWS Bedrock	0.230	0.380	compare	128K	8K	10.3#185	9.6#139
Claude 3 Sonnet 20240229 V1	AWS Bedrock	3.00	15.00	compare	200K	4K	10.3#185	N/A
Nova Micro V1	AWS Bedrock	0.035	0.140	compare	128K	10K	10.3#185	4.1#160
Nvidia.nemotron Nano 12B V2	AWS Bedrock	0.200	0.600	compare	128K	8K	10.1#189	5.9#155
Llama3 2 3B Instruct V1	AWS Bedrock	0.150	0.150	compare	128K	4K	9.7#196	N/A
Claude V2	AWS Bedrock	8.00	24.00	compare	100K	8K	9.3#199	14.0#105
Mistral Small 2402 V1	AWS Bedrock	1.00	3.00	compare	32K	8K	9.0#201	N/A
Llama3 70B Instruct V1	AWS Bedrock	2.65	3.50	compare	8K	8K	8.9#204	6.8#151
Gemma 3 12B It	AWS Bedrock	0.090	0.290	compare	128K	8K	8.8#205	6.3#154
Llama3 2 11B Instruct V1	AWS Bedrock	0.350	0.350	compare	128K	4K	8.7#209	4.3#159
Llama2 70B Chat V1	AWS Bedrock	1.95	2.56	compare	4K	4K	8.4#212	N/A
Llama2 13B Chat V1	AWS Bedrock	0.750	1.00	compare	4K	4K	8.4#212	N/A
Command R Plus V1	AWS Bedrock	3.00	15.00	compare	128K	4K	8.3#216	N/A
Jamba 1 5 Mini V1	AWS Bedrock	0.200	0.400	compare	256K	256K	8.0#221	N/A
Command R V1	AWS Bedrock	0.500	1.50	compare	128K	4K	7.4#227	N/A
Claude Instant V1	AWS Bedrock	0.800	2.40	compare	100K	8K	7.4#227	7.8#142
Llama3 8B Instruct V1	AWS Bedrock	0.300	0.600	compare	8K	8K	6.4#233	4.0#161
Llama3 2 1B Instruct V1	AWS Bedrock	0.100	0.100	compare	128K	4K	6.3#234	0.6#173
Gemma 3 4B It	AWS Bedrock	0.040	0.080	compare	128K	8K	6.3#234	2.9#167
Us.writer.palmyra X5 V1	AWS Bedrock	0.600	6.00	compare	1.0M	8K	N/A	N/A
Us.writer.palmyra X4 V1	AWS Bedrock	2.50	10.00	compare	128K	8K	N/A	N/A
GPT-oss-safeguard-20b	AWS Bedrock	0.070	0.200	compare	128K	8K	N/A	N/A
GPT-oss-safeguard-120b	AWS Bedrock	0.150	0.600	compare	128K	8K	N/A	N/A
Nvidia.nemotron Super 3 120B	AWS Bedrock	0.150	0.650	compare	256K	33K	N/A	N/A
Voxtral Small 24B 2507	AWS Bedrock	0.100	0.300	compare	128K	8K	N/A	N/A
Voxtral Mini 3B 2507	AWS Bedrock	0.040	0.040	compare	128K	8K	N/A	N/A
Devstral 2 123B	AWS Bedrock	0.400	2.00	compare	256K	8K	N/A	N/A
Llama4 Scout 17B Instruct V1	AWS Bedrock	0.170	0.660	compare	128K	4K	N/A	N/A
Llama4 Maverick 17B Instruct V1	AWS Bedrock	0.240	0.970	compare	128K	4K	N/A	N/A
Eu.twelvelabs.pegasus 1 2 V1	AWS Bedrock	N/A	7.50	compare	N/A	N/A	N/A	N/A
Eu.mistral.pixtral Large 2502 V1	AWS Bedrock	2.00	6.00	compare	128K	4K	N/A	N/A
Mixtral 8x7B Instruct V0	AWS Bedrock	0.450	0.700	compare	32K	8K	N/A	N/A
Mistral Large 2402 V1	AWS Bedrock	8.00	24.00	compare	32K	8K	N/A	N/A
Mistral 7B Instruct V0	AWS Bedrock	0.150	0.200	compare	32K	8K	N/A	N/A
Qwen3 Coder Next	AWS Bedrock	0.600	1.44	compare	262K	8K	N/A	N/A
Minimax.minimax M2.1	AWS Bedrock	0.360	1.44	compare	196K	8K	N/A	N/A
Command Text V14	AWS Bedrock	N/A	N/A	compare	4K	4K	N/A	N/A
Command Light Text V14	AWS Bedrock	N/A	N/A	compare	4K	4K	N/A	N/A
Claude V1	AWS Bedrock	8.00	24.00	compare	100K	8K	N/A	N/A
Titan Text Premier V1	AWS Bedrock	0.500	1.50	compare	42K	32K	N/A	N/A
Titan Text Lite V1	AWS Bedrock	0.300	0.400	compare	42K	4K	N/A	N/A
Titan Text Express V1	AWS Bedrock	1.30	1.70	compare	42K	8K	N/A	N/A
Jamba Instruct V1	AWS Bedrock	0.500	0.700	compare	70K	4K	N/A	N/A
J2 Ultra V1	AWS Bedrock	18.80	18.80	compare	8K	8K	N/A	N/A
J2 Mid V1	AWS Bedrock	12.50	12.50	compare	8K	8K	N/A	N/A

AI Models Comparison