Amazon Bedrock: Models Intelligence, Performance & Price

Name: Intelligence
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

Amazon Bedrock

Analysis of Amazon Bedrock's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Amazon Bedrock for your use-case.

Most Intelligent

Updated

Claude Opus 4.8 (max)

GPT-5.5 (xhigh)

Claude Opus 4.7 (max)

Claude Sonnet 5 (max)

GPT-5.5 (high)

Intelligence index

Total 91 models

Fastest

Ministral 3 3B

411 t/s

Nova Micro

283 t/s

gpt-oss-20b (high)

281 t/s

gpt-oss-20b (low)

277 t/s

Ministral 3 14B

237 t/s

Output speed

Total 91 models

Lowest Price

Nova Micro

$0.03

Gemma 3 4B

$0.04

Nova Lite

$0.05

NVIDIA Nemotron Nano 9B V2

$0.08

gpt-oss-20b (high)

$0.09

Blended price (per 1M tokens)

Total 91 models

Indicates a reasoning model

Amazon offers 91 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.

For intelligence, the top models on Amazon are Claude Opus 4.8 (max) (56), GPT-5.5 (xhigh) (55), Claude Opus 4.7 (max) (54).
For output speed, the fastest models are Ministral 3 3B (411 t/s), Nova Micro (283 t/s), gpt-oss-20b (high) (281 t/s). Speed varies significantly across models, with a 73% difference between the fastest and slowest.
For latency, Llama 3.2 11B (Vision) (0.73s), Llama 3.1 8B (0.76s), Llama 3.2 1B (0.85s) offer the lowest time to first token.
For pricing, Nova Micro ($0.03), Gemma 3 4B ($0.04), Nova Lite ($0.05) offer the lowest blended prices per 1M tokens. Prices vary up to 3.4x across models.
For context window size, Claude Opus 4.8 (max) (1M), Claude Opus 4.7 (max) (1M), Claude Sonnet 5 (max) (1M) support the largest context windows on Amazon.

Highlights

Updated

Intelligence

Artificial Analysis Intelligence Index · Higher is better

Speed

Output tokens per second · Higher is better

Price

USD per 1M tokens (blended) · Lower is better

Intelligence Evaluations

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR

Estimate (independent evaluation forthcoming)

Reasoning models are indicated by a lightbulb icon

Artificial Analysis Intelligence Index v4.1 includes: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

GDPval-AA v2

Agentic real-world work tasks, (Elo-500)/2000

𝜏³-Banking

Agentic tool use

Terminal-Bench v2.1

Agentic coding & terminal use

SciCode

Coding

Humanity's Last Exam

Reasoning & knowledge

GPQA Diamond

Scientific reasoning

CritPt

Physics reasoning

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AA-LCR

Long context reasoning

AA-BriefcaseNew

Agentic knowledge work, Elo

AutomationBench-AANew

Agentic SaaS workflows

Harvey LAB-AANew

Legal agentic work, task all-pass rate

EnterpriseOps-Gym-AANew

Agentic business operations

IFBench

Instruction following

APEX-Agents-AA

Long-horizon agentic tasks

ITBench-AA

Kubernetes incident root-cause analysis

MMMU-Pro

Visual reasoning

Reasoning models are indicated by a lightbulb icon.

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

Intelligence vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon.

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive).
Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Context Window

Context window: tokens limit · Higher is better

Reasoning models are indicated by a lightbulb icon

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Pricing

Intelligence vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon.

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive).
Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Performance Summary

Output Speed vs. Price

Output speed: output tokens per second · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon.

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Speed

Measured by Output Speed (tokens per second)

Output Speed

Output tokens per second · Higher is better

Reasoning models are indicated by a lightbulb icon

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Latency

Measured by Time (seconds) to First Token

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time

Reasoning models are indicated by a lightbulb icon

Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.

End-to-End Response Time

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

End-to-End Response Time vs. Price

End-to-end response time: end-to-end seconds to output 500 tokens · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Seconds to receive a 500 token response. Key components:

Input time: Time to receive the first response token
Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
Answer time: Time to generate 500 output tokens, based on output speed

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.


									Further Analysis
Claude Opus 4.8 (max)	1M	Proprietary	56	$3.85	63	26.87	34.82	--	Model Provider
GPT-5.5 (xhigh)	272k	Proprietary	55	$4.79	129	53.46	57.34	--	Model Provider
Claude Opus 4.7 (max)	1M	Proprietary	54	$3.85	69	1.95	9.23	--	Model Provider
Claude Sonnet 5 (max)	1M	Proprietary	53	$2.31	77	226.58	233.06	--	Model Provider
GPT-5.5 (high)	272k	Proprietary	53	$4.79	128	16.87	20.78	--	Model Provider
GPT-5.4 (xhigh)	272k	Proprietary	51	$2.39	39	266.64	279.38	--	Model Provider
GPT-5.5 (medium)	272k	Proprietary	50	$4.79	115	6.63	10.99	--	Model Provider
Claude Sonnet 4.6 (max)	200k	Proprietary	47	$2.31	71	145.09	152.17	--	Model Provider
Claude Opus 4.6 (max)	1M	Proprietary	44*	$3.85	45	11.91	23.00	--	Model Provider
GPT-5.5 (low)	272k	Proprietary	43	$4.79	121	2.77	6.92	--	Model Provider
Claude Opus 4.7 (Non-reasoning, high)	1M	Proprietary	43*	$3.85	65	1.77	9.48	--	Model Provider
Claude Opus 4.5	200k	Proprietary	41*	$3.85	45	13.77	24.93	--	Model Provider
GPT-5.4 (low)	272k	Proprietary	39*	$2.39	37	5.20	18.85	--	Model Provider
Claude Opus 4.6 (high)	1M	Proprietary	38*	$3.85	43	1.89	13.50	--	Model Provider
Grok 4.3 (high)	524k	Proprietary	38	$0.64	222	15.83	18.08	--	Model Provider
Claude 4.5 Sonnet	1M	Proprietary	36	$2.31	48	10.69	21.19	--	Model Provider
Grok 4.3 (medium)	524k	Proprietary	36*	$0.64	196	9.17	11.72	--	Model Provider
Claude Sonnet 4.6 (Non-reasoning)	200k	Proprietary	36*	$2.31	45	1.39	12.54	--	Model Provider
Grok 4.3 (low)	524k	Proprietary	35*	$0.64	198	3.99	6.52	--	Model Provider
GPT-5.5 (Non-reasoning)	272k	Proprietary	35	$4.79	106	1.10	5.83	--	Model Provider
Kimi K2.5	256k	Open	35	$0.84	111	1.30	32.57	26.76	Model Provider
Claude Opus 4.5	200k	Proprietary	35*	$3.85	44	1.58	12.94	--	Model Provider
Claude 4.1 Opus	200k	Proprietary	34*	$11.55	15	33.72	66.30	--	Model Provider
GLM-4.7	200k	Open	34	$0.76	59	1.49	43.53	33.64	Model Provider
Kimi K2 Thinking	256k	Open	33*	$0.79	121	1.44	22.14	16.56	Model Provider
DeepSeek V3.2	128k	Open	32	$0.74	85	1.54	30.90	23.49	Model Provider
Claude 4.5 Haiku	200k	Proprietary	30	$0.77	108	10.62	15.24	--	Model Provider
Claude 4.5 Sonnet	1M	Proprietary	29*	$2.31	45	1.74	12.90	--	Model Provider
Claude 4 Sonnet	1M	Proprietary	29	$2.31	--	--	--	--	Model Provider
MiniMax-M2	205k	Open	28*	$0.39	119	1.06	22.01	16.76	Model Provider
Claude 4.1 Opus	200k	Proprietary	28*	$11.55	14	4.31	40.25	--	Model Provider
GPT-5.4 (Non-reasoning)	272k	Proprietary	28*	$2.39	33	2.10	17.36	--	Model Provider
GLM-4.7	200k	Open	27*	$0.76	62	1.58	9.63	--	Model Provider
Claude 4 Sonnet	1M	Proprietary	25*	$2.31	62	1.64	9.73	--	Model Provider
Grok 4.3 (Non-reasoning)	524k	Proprietary	25	$0.64	194	1.01	3.59	--	Model Provider
DeepSeek V3.2	128k	Open	25*	$0.74	74	1.72	8.49	--	Model Provider
gpt-oss-120b (high)	131k	Open	24	$0.20	121	1.26	21.91	16.53	Model Provider
Claude 4.5 Haiku	200k	Proprietary	24*	$0.77	105	1.05	5.81	--	Model Provider
GLM-4.7-Flash	200k	Open	23*	$0.10	208	1.17	13.21	9.63	Model Provider
Nova 2.0 Pro Preview (medium)	256k	Proprietary	22	$1.47	122	12.36	32.83	16.38	Model Provider
Qwen3 Coder Next	128k	Open	21	$0.57	138	2.60	6.23	--	Model Provider
DeepSeek V3.1	128k	Open	21*	$0.69	177	1.43	4.25	--	Model Provider
Nova 2.0 Omni (medium)	1M	Proprietary	21*	$0.52	--	--	--	--	Model Provider
Nova 2.0 Omni (medium)	1M	Proprietary	21*	$0.52	--	--	--	--	Model Provider
DeepSeek V3.1	128k	Open	21*	$0.69	161	1.43	16.99	12.45	Model Provider
Nova 2.0 Pro Preview (low)	256k	Proprietary	20	$2.13	126	13.05	32.83	15.83	Model Provider
Nova 2.0 Lite (medium)	1M	Proprietary	19*	$0.52	156	18.37	34.42	12.84	Model Provider
DeepSeek R1 (Jan)	128k	Open	19	$1.76	145	1.22	20.89	16.21	Model Provider
Nova 2.0 Lite (high)	1M	Proprietary	18	$0.52	--	--	--	--	Model Provider
Nova 2.0 Lite (high)	1M	Proprietary	18	$0.52	156	17.62	33.62	12.80	Model Provider
Qwen3 235B 2507	256k	Open	18*	$0.29	75	1.40	8.06	--	Model Provider
Qwen3 Coder 480B	262k	Open	18*	$0.38	115	1.30	5.65	--	Model Provider
Nova 2.0 Lite (low)	1M	Proprietary	18*	$0.52	147	9.50	26.55	13.64	Model Provider
Nova 2.0 Omni (low)	1M	Proprietary	17*	$0.52	--	--	--	--	Model Provider
Mistral Large 3	256k	Open	16	$0.60	169	1.22	4.17	--	Model Provider
GLM-4.7-Flash	200k	Open	16*	$0.10	207	1.21	3.63	--	Model Provider
gpt-oss-120b (low)	131k	Open	15	$0.20	97	1.13	26.99	20.69	Model Provider
gpt-oss-20b (high)	131k	Open	15	$0.09	281	20.60	29.51	7.12	Model Provider
Nova 2.0 Pro Preview	256k	Proprietary	14	$2.13	110	1.06	5.61	--	Model Provider
gpt-oss-20b (low)	131k	Open	14*	$0.09	277	6.32	15.34	7.22	Model Provider
Llama 4 Maverick	128k	Open	14	$0.31	214	0.87	3.20	--	Model Provider
Qwen3 Coder 30B A3B	262k	Open	14*	$0.20	220	1.12	3.39	--	Model Provider
Nova Premier	1M	Proprietary	13*	$2.18	32	2.87	18.37	--	Model Provider
Nova 2.0 Lite	1M	Proprietary	12*	$0.52	142	1.05	4.57	--	Model Provider
Magistral Small 1.2	128k	Open	11	$0.60	74	1.58	35.39	27.06	Model Provider
Ministral 3 14B	256k	Open	11	$0.20	237	1.00	3.10	--	Model Provider
Nova 2.0 Omni	1M	Proprietary	11*	$0.52	--	--	--	--	Model Provider
Llama 4 Scout	128k	Open	10	$0.22	157	0.87	4.05	--	Model Provider
Claude 3.5 Sonnet (Oct)	200k	Proprietary	10*	$2.31	--	--	--	--	Model Provider
Llama 3.3 70B	128k	Open	9	$0.71	125	1.26	5.27	--	Model Provider
Ministral 3 8B	256k	Open	9	$0.15	217	1.05	3.35	--	Model Provider
Qwen3 32B	32.8k	Open	9*	$0.20	159	1.12	4.27	--	Model Provider
Llama 3.1 405B Standard	128k	Open	9*	$2.40	--	--	--	--	Model Provider
Llama 3.1 405B Latency Optimized	128k	Open	9*	$3.00	79	2.44	8.80	--	Model Provider
Claude 3.5 Sonnet (June)	200k	Proprietary	8*	$2.31	--	--	--	--	Model Provider
Nova Pro	300k	Proprietary	8*	$0.62	120	1.18	5.35	--	Model Provider
Llama 3.1 8B	128k	Open	8	$0.22	185	0.76	3.46	--	Model Provider
Gemma 3 27B	128k	Open	7	$0.25	60	1.44	9.81	--	Model Provider
NVIDIA Nemotron Nano 9B V2	131k	Open	7*	$0.08	170	1.18	4.12	--	Model Provider
Mistral Large 2 (Jul)	128k	Open	7*	$2.40	29	1.86	18.82	--	Model Provider
Nova Lite	300k	Proprietary	7*	$0.05	172	0.99	3.89	--	Model Provider
Ministral 3 3B	256k	Open	7	$10.00	411	0.95	2.16	--	Model Provider
Llama 3.1 70B Standard	128k	Open	7*	$0.72	14	42.24	77.63	--	Model Provider
Llama 3.1 70B Latency Optimized	128k	Open	7*	$0.90	122	1.28	5.37	--	Model Provider
Llama 3.2 90B (Vision)	128k	Open	6*	$0.72	57	1.11	9.83	--	Model Provider
Gemma 3 12B	128k	Open	6	$0.11	96	1.40	6.60	--	Model Provider
Jamba 1.5 Large	256k	Open	5*	$2.60	46	2.90	13.84	--	Model Provider
Claude 3 Sonnet	200k	Proprietary	5*	$4.20	45	3.05	14.11	--	Model Provider
Nova Micro	130k	Proprietary	5*	$0.03	283	1.09	2.86	--	Model Provider
NVIDIA Nemotron Nano 12B v2 VL	128k	Open	5*	$0.24	206	1.16	3.58	--	Model Provider
Mistral Large (Feb)	32.8k	Proprietary	4*	$4.80	41	1.56	13.66	--	Model Provider
Llama 3.2 3B	128k	Open	4*	$0.15	52	1.07	10.70	--	Model Provider
Claude 3 Haiku	200k	Proprietary	4*	$0.35	81	1.02	7.18	--	Model Provider
Llama 3 70B	8.19k	Open	3*	$2.74	--	--	--	--	Model Provider
Llama 3.2 11B (Vision)	128k	Open	3*	$0.16	47	0.73	11.40	--	Model Provider
Command-R+ (Apr)	128k	Open	3*	$4.20	37	2.52	16.03	--	Model Provider
Jamba 1.5 Mini	256k	Open	3*	$0.22	52	4.59	14.18	--	Model Provider
Mixtral 8x7B	32.8k	Open	2*	$0.47	25	1.36	21.67	--	Model Provider
Mistral 7B	8.19k	Open	2*	$0.15	--	--	--	--	Model Provider
Command-R (Mar)	128k	Open	2*	$0.60	70	1.34	8.47	--	Model Provider
Llama 3 8B	8.19k	Open	1*	$0.33	--	--	--	--	Model Provider
Gemma 3 4B	128k	Open	1*	$0.04	226	0.96	3.18	--	Model Provider
Llama 3.2 1B	128k	Open	1*	$0.10	86	0.85	6.67	--	Model Provider

Key definitions

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Price per token generated by the model (received from the API), represented as USD per million Tokens.

Price per token included in the request/message sent to the API, represented as USD per million Tokens.

Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.

Frequently Asked Questions

Common questions about Amazon Bedrock

Amazon Bedrock offers 91 models that we track: Claude Opus 4.8 (max), GPT-5.5 (xhigh), Claude Opus 4.7 (max), Claude Sonnet 5 (max), GPT-5.5 (high), GPT-5.4 (xhigh), GPT-5.5 (medium), Claude Sonnet 4.6 (max), Claude Opus 4.6 (max), GPT-5.5 (low), Claude Opus 4.7 (Non-reasoning, high), Claude Opus 4.5, GPT-5.4 (low), Claude Opus 4.6 (high), Grok 4.3 (high), Claude 4.5 Sonnet, Grok 4.3 (medium), Claude Sonnet 4.6 (Non-reasoning), Grok 4.3 (low), GPT-5.5 (Non-reasoning), Kimi K2.5, Claude Opus 4.5, Claude 4.1 Opus, GLM-4.7, Kimi K2 Thinking, DeepSeek V3.2, Claude 4.5 Haiku, Claude 4.5 Sonnet, MiniMax-M2, Claude 4.1 Opus, GPT-5.4 (Non-reasoning), GLM-4.7, Claude 4 Sonnet, Grok 4.3 (Non-reasoning), DeepSeek V3.2, gpt-oss-120b (high), Claude 4.5 Haiku, GLM-4.7-Flash, Nova 2.0 Pro Preview (medium), Qwen3 Coder Next, DeepSeek V3.1, DeepSeek V3.1, Nova 2.0 Pro Preview (low), Nova 2.0 Lite (medium), DeepSeek R1 (Jan), Nova 2.0 Lite (high), Qwen3 235B 2507, Qwen3 Coder 480B, Nova 2.0 Lite (low), Mistral Large 3, GLM-4.7-Flash, gpt-oss-120b (low), gpt-oss-20b (high), Nova 2.0 Pro Preview, gpt-oss-20b (low), Llama 4 Maverick, Qwen3 Coder 30B A3B, Nova Premier, Nova 2.0 Lite, Magistral Small 1.2, Ministral 3 14B, Llama 4 Scout, Llama 3.3 70B, Ministral 3 8B, Qwen3 32B, Llama 3.1 405B Latency Optimized, Nova Pro, Llama 3.1 8B, Gemma 3 27B, NVIDIA Nemotron Nano 9B V2, Mistral Large 2 (Jul), Nova Lite, Ministral 3 3B, Llama 3.1 70B Standard, Llama 3.1 70B Latency Optimized, Llama 3.2 90B (Vision), Gemma 3 12B, Jamba 1.5 Large, Claude 3 Sonnet, Nova Micro, NVIDIA Nemotron Nano 12B v2 VL, Mistral Large (Feb), Llama 3.2 3B, Claude 3 Haiku, Llama 3.2 11B (Vision), Command-R+ (Apr), Jamba 1.5 Mini, Mixtral 8x7B, Command-R (Mar), Gemma 3 4B, and Llama 3.2 1B.

The most intelligent model available on Amazon Bedrock is Claude Opus 4.8 (max) with an Intelligence Index score of 56.

The fastest model on Amazon Bedrock by output speed is Ministral 3 3B at 411.3 tokens per second.

The model with the lowest time to first token on Amazon Bedrock is Llama 3.2 11B (Vision) at 0.73s. Lower latency means faster initial response time.

The most affordable model on Amazon Bedrock by blended price is Nova Micro at $0.03 per 1M tokens (7:2:1 cache hit/input/output ratio).

Prices on Amazon Bedrock vary up to 426x across models, from $0.03 per 1M tokens for Nova Micro to $11.55 per 1M tokens for Claude 4.1 Opus.

18 of 91 models on Amazon Bedrock support JSON mode for structured output.

87 of 91 models on Amazon Bedrock support function calling (tool use).

Yes, Amazon Bedrock offers 36 reasoning models: Claude Opus 4.8 (max), GPT-5.5 (xhigh), Claude Opus 4.7 (max), Claude Sonnet 5 (max), GPT-5.5 (high), GPT-5.4 (xhigh), GPT-5.5 (medium), Claude Sonnet 4.6 (max), Claude Opus 4.6 (max), GPT-5.5 (low), Claude Opus 4.5, GPT-5.4 (low), Grok 4.3 (high), Claude 4.5 Sonnet, Grok 4.3 (medium), Grok 4.3 (low), Kimi K2.5, Claude 4.1 Opus, GLM-4.7, Kimi K2 Thinking, DeepSeek V3.2, Claude 4.5 Haiku, MiniMax-M2, gpt-oss-120b (high), GLM-4.7-Flash, Nova 2.0 Pro Preview (medium), DeepSeek V3.1, Nova 2.0 Pro Preview (low), Nova 2.0 Lite (medium), DeepSeek R1 (Jan), Nova 2.0 Lite (high), Nova 2.0 Lite (low), gpt-oss-120b (low), gpt-oss-20b (high), gpt-oss-20b (low), and Magistral Small 1.2. Reasoning models use extended thinking to work through complex problems before providing an answer.

Yes, 48 of 91 models on Amazon Bedrock are open weight models: Kimi K2.5, GLM-4.7, Kimi K2 Thinking, DeepSeek V3.2, MiniMax-M2, GLM-4.7, DeepSeek V3.2, gpt-oss-120b (high), GLM-4.7-Flash, Qwen3 Coder Next, DeepSeek V3.1, DeepSeek V3.1, DeepSeek R1 (Jan), Qwen3 235B 2507, Qwen3 Coder 480B, Mistral Large 3, GLM-4.7-Flash, gpt-oss-120b (low), gpt-oss-20b (high), gpt-oss-20b (low), Llama 4 Maverick, Qwen3 Coder 30B A3B, Magistral Small 1.2, Ministral 3 14B, Llama 4 Scout, Llama 3.3 70B, Ministral 3 8B, Qwen3 32B, Llama 3.1 405B Latency Optimized, Llama 3.1 8B, Gemma 3 27B, NVIDIA Nemotron Nano 9B V2, Mistral Large 2 (Jul), Ministral 3 3B, Llama 3.1 70B Standard, Llama 3.1 70B Latency Optimized, Llama 3.2 90B (Vision), Gemma 3 12B, Jamba 1.5 Large, NVIDIA Nemotron Nano 12B v2 VL, Llama 3.2 3B, Llama 3.2 11B (Vision), Command-R+ (Apr), Jamba 1.5 Mini, Mixtral 8x7B, Command-R (Mar), Gemma 3 4B, and Llama 3.2 1B.

Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.

When choosing a model on Amazon Bedrock, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.

Amazon Bedrock: Models Intelligence, Performance & Price

Most Intelligent

Fastest

Lowest Price

Intelligence