Does Google offer open weight models?

Yes, 20 of 52 models on Google are open weight models: GLM-5, GLM-4.7, Kimi K2 Thinking Vertex, Gemma 4 31B (AI Studio), MiniMax-M2 Vertex, GLM-4.7, Gemma 4 26B A4B AI Studio, gpt-oss-120b (high) Vertex, DeepSeek V3.1 Vertex, DeepSeek V3.1 Vertex, DeepSeek R1 0528 Vertex, Qwen3 235B 2507 Vertex, Qwen3 Coder 480B Vertex, Qwen3 Next 80B A3B Vertex, gpt-oss-120b (low) Vertex, gpt-oss-20b (high) Vertex, gpt-oss-20b (low) Vertex, Qwen3 Next 80B A3B Vertex, Llama 4 Scout Vertex, and Llama 3.3 70B Vertex.

Artificial Analysis

Google: Models Intelligence, Performance & Price

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

Google

Analysis of Google'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 Google for your use-case.

Most Intelligent

Updated

Claude Fable 5 (with fallback)

Claude Opus 4.8 (max)

Claude Opus 4.7 (max)

Claude Sonnet 5 (max)

Gemini 3.5 Flash AI Studio

Intelligence index

Total 52 models

Fastest

Gemini 3.5 Flash-Lite AI Studio

463 t/s

gpt-oss-120b (low) Vertex

414 t/s

gpt-oss-120b (high) Vertex

397 t/s

Gemini 3.1 Flash-Lite (AI Studio)

303 t/s

Gemini 2.5 Flash-Lite (AI Studio)

290 t/s

Output speed

Total 52 models

Lowest Price

Gemini 2.5 Flash-Lite (AI Studio)

$0.07

Gemini 2.5 Flash-Lite (AI Studio)

$0.07

gpt-oss-20b (high) Vertex

$0.09

gpt-oss-20b (low) Vertex

$0.09

gpt-oss-120b (high) Vertex

$0.12

Blended price (per 1M tokens)

Total 52 models

Indicates a reasoning model

Google offers 52 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 Google are Claude Fable 5 (with fallback) (60), Claude Opus 4.8 (max) (56), Claude Opus 4.7 (max) (54).
For output speed, the fastest models are Gemini 3.5 Flash-Lite AI Studio (463 t/s), gpt-oss-120b (low) Vertex (414 t/s), gpt-oss-120b (high) Vertex (397 t/s). Speed varies significantly across models, with a 60% difference between the fastest and slowest.
For latency, Gemini 2.5 Flash-Lite (AI Studio) (0.33s), Gemini 2.5 Flash (AI Studio) (0.51s), Claude 4.5 Haiku Vertex (0.59s) offer the lowest time to first answer token.
For pricing, Gemini 2.5 Flash-Lite (AI Studio) ($0.07), Gemini 2.5 Flash-Lite (AI Studio) ($0.07), gpt-oss-20b (high) Vertex ($0.09) offer the lowest blended prices per 1M tokens.
For context window size, Llama 4 Scout Vertex (1M), Claude Fable 5 (with fallback) (1M), Claude Opus 4.8 (max) (1M) support the largest context windows on Google.

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, criterion 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 Index 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 Index 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 Fable 5 (with fallback)	1M	Proprietary	60	$2.90	73	108.85	115.66	--	Model Provider
Claude Opus 4.8 (max)	1M	Proprietary	56	$2.39	68	32.92	40.26	--	Model Provider
Claude Opus 4.7 (max)	1M	Proprietary	54	$3.37	71	14.97	22.00	--	Model Provider
Claude Sonnet 5 (max)	1M	Proprietary	53	$2.92	73	199.94	206.76	--	Model Provider
Gemini 3.5 Flash AI Studio	1M	Proprietary	50	$0.59	183	17.98	20.71	--	Model Provider
Gemini 3.6 Flash AI Studio	1M	Proprietary	50	$0.50	251	13.98	15.97	--	Model Provider
Claude Sonnet 4.6 (max)	200k	Proprietary	47	$1.63	64	119.83	127.68	--	Model Provider
Gemini 3.1 Pro Preview (AI Studio)	1M	Proprietary	46	$0.29	116	28.19	32.51	--	Model Provider
Gemini 3.1 Pro Preview (Vertex)	1M	Proprietary	46	$0.36	123	32.16	36.23	--	Model Provider
Gemini 3.5 Flash (medium)	1M	Proprietary	45*	--	192	20.64	23.25	--	Model Provider
Claude Opus 4.6 (max)	1M	Proprietary	44*	--	44	16.20	27.45	--	Model Provider
Claude Opus 4.5 Vertex	200k	Proprietary	41*	--	50	13.09	23.10	--	Model Provider
Gemini 3 Pro Preview (high) (Vertex)	1M	Proprietary	40*	--	--	--	--	--	Model Provider
GLM-5	200k	Open	40*	--	166	1.01	22.67	18.66	Model Provider
Claude Opus 4.6 (high)	1M	Proprietary	38*	--	42	1.76	13.76	--	Model Provider
Gemini 3 Flash (AI Studio)	1M	Proprietary	38*	--	170	7.07	10.02	--	Model Provider
Gemini 3.5 Flash-Lite AI Studio	1M	Proprietary	36	$0.09	463	8.88	9.96	--	Model Provider
Claude 4.5 Sonnet Vertex	1M	Proprietary	36	$0.40	47	12.47	23.01	--	Model Provider
Claude Sonnet 4.6 (Non-reasoning)	200k	Proprietary	36*	--	44	1.13	12.53	--	Model Provider
Gemini 3.5 Flash (minimal) AI Studio	1M	Proprietary	35*	--	158	1.01	4.17	--	Model Provider
Claude Opus 4.5 Vertex	200k	Proprietary	35*	--	45	1.02	12.04	--	Model Provider
Claude 4.1 Opus Vertex	200k	Proprietary	34*	--	--	--	--	--	Model Provider
GLM-4.7	200k	Open	34	--	148	0.70	17.61	13.53	Model Provider
Gemini 3 Pro Preview (low) (Vertex)	1M	Proprietary	33*	--	--	--	--	--	Model Provider
Kimi K2 Thinking Vertex	262k	Open	33*	--	189	0.81	14.02	10.57	Model Provider
Claude 4 Opus Vertex	200k	Proprietary	31*	--	--	--	--	--	Model Provider
Claude 4.5 Haiku Vertex	200k	Proprietary	30	$0.23	99	12.82	17.86	--	Model Provider
Gemma 4 31B (AI Studio)	262k	Open	29	--	35	1.11	64.59	49.28	Model Provider
Claude 4.5 Sonnet Vertex	1M	Proprietary	29*	--	39	1.15	13.92	--	Model Provider
Claude 4 Sonnet Vertex	1M	Proprietary	29	$0.45	--	--	--	--	Model Provider
MiniMax-M2 Vertex	197k	Open	28*	--	148	0.63	17.56	13.54	Model Provider
Claude 4.1 Opus Vertex	200k	Proprietary	28*	--	--	--	--	--	Model Provider
Gemini 3 Flash (AI Studio)	1M	Proprietary	27*	--	168	0.95	3.93	--	Model Provider
GLM-4.7	200k	Open	27*	--	142	0.70	4.21	--	Model Provider
Gemini 2.5 Pro Vertex	1M	Proprietary	26	$0.20	112	31.87	36.32	--	Model Provider
Gemini 2.5 Pro (AI Studio)	1M	Proprietary	26	$0.20	135	20.80	24.50	--	Model Provider
Gemma 4 26B A4B AI Studio	262k	Open	26	--	47	1.13	54.39	42.61	Model Provider
Claude 4 Opus Vertex	200k	Proprietary	25*	--	--	--	--	--	Model Provider
Claude 4 Sonnet Vertex	1M	Proprietary	25*	--	--	--	--	--	Model Provider
Gemini 3.1 Flash-Lite (AI Studio)	1M	Proprietary	25	$0.04	303	6.38	8.03	--	Model Provider
gpt-oss-120b (high) Vertex	131k	Open	24	--	397	0.49	6.79	5.04	Model Provider
Claude 4.5 Haiku Vertex	200k	Proprietary	24*	--	92	0.59	6.02	--	Model Provider
Claude 3.7 Sonnet Vertex	200k	Proprietary	24*	--	--	--	--	--	Model Provider
Gemini 2.5 Pro (May) (AI Studio)	1M	Proprietary	22*	--	--	--	--	--	Model Provider
DeepSeek V3.1 Vertex	164k	Open	21*	--	163	0.83	3.90	--	Model Provider
DeepSeek V3.1 Vertex	164k	Open	21*	--	147	0.86	17.83	13.57	Model Provider
DeepSeek R1 0528 Vertex	164k	Open	20*	--	156	0.98	17.03	12.84	Model Provider
Gemini 2.5 Flash (AI Studio)	1M	Proprietary	20*	--	231	18.02	20.18	--	Model Provider
Gemini 2.5 Flash (Vertex)	1M	Proprietary	20*	--	186	18.85	21.54	--	Model Provider
Qwen3 235B 2507 Vertex	262k	Open	18*	--	81	0.83	7.02	--	Model Provider
Qwen3 Coder 480B Vertex	262k	Open	18*	--	93	4.98	10.38	--	Model Provider
Qwen3 Next 80B A3B Vertex	262k	Open	17	--	130	0.80	19.99	15.35	Model Provider
Gemini 2.5 Flash-Lite (Sep) (AI Studio)	1M	Proprietary	15*	--	--	--	--	--	Model Provider
gpt-oss-120b (low) Vertex	131k	Open	15	--	414	0.50	6.54	4.83	Model Provider
gpt-oss-20b (high) Vertex	131k	Open	15	--	289	1.19	9.84	6.92	Model Provider
gpt-oss-20b (low) Vertex	131k	Open	14*	--	244	0.84	11.11	8.21	Model Provider
Gemini 2.5 Flash (Vertex)	1M	Proprietary	14*	--	166	0.78	3.79	--	Model Provider
Gemini 2.5 Flash (AI Studio)	1M	Proprietary	14*	--	203	0.51	2.97	--	Model Provider
Qwen3 Next 80B A3B Vertex	262k	Open	14*	--	196	0.71	3.26	--	Model Provider
Gemini 2.5 Flash-Lite (Sep) (AI Studio)	1M	Proprietary	13*	--	--	--	--	--	Model Provider
Gemini 2.5 Flash-Lite (AI Studio)	1M	Proprietary	11*	--	290	24.59	26.32	--	Model Provider
Gemini 2.0 Flash (exp) (AI Studio)	1M	Proprietary	11*	--	--	--	--	--	Model Provider
Llama 4 Scout Vertex	1.31M	Open	10	$0.01	143	0.84	4.33	--	Model Provider
Llama 3.3 70B Vertex	128k	Open	9	$0.10	148	0.66	4.04	--	Model Provider
Gemma 3 27B (AI Studio)	128k	Open	7	--	--	--	--	--	Model Provider
Gemini 2.5 Flash-Lite (AI Studio)	1M	Proprietary	7*	--	240	0.33	2.42	--	Model Provider
Gemma 3 12B (AI Studio)	128k	Open	6	--	--	--	--	--	Model Provider
Gemma 3 4B (AI Studio)	128k	Open	1*	--	--	--	--	--	Model Provider
Gemma 3n E2B (AI Studio)	32k	Open	1*	--	--	--	--	--	Model Provider
Gemma 3 1B (AI Studio)	32k	Open	1*	--	--	--	--	--	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.

Average cost per task in the index. Costs are split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available.

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

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Price per token to write prompt tokens into the cache so that later requests can hit them, represented as USD per million tokens. Some providers charge a premium over the standard input price to create a cache entry (e.g. Anthropic), while others cache automatically with no separate write fee.

Price per token generated by the model (received from 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 Google

Google offers 52 models that we track: Claude Fable 5 (with fallback), Claude Opus 4.8 (max), Claude Opus 4.7 (max), Claude Sonnet 5 (max), Gemini 3.5 Flash AI Studio, Gemini 3.6 Flash AI Studio, Claude Sonnet 4.6 (max), Gemini 3.1 Pro Preview (AI Studio), Gemini 3.1 Pro Preview (Vertex), Gemini 3.5 Flash (medium), Claude Opus 4.6 (max), Claude Opus 4.5 Vertex, GLM-5, Claude Opus 4.6 (high), Gemini 3 Flash (AI Studio), Gemini 3.5 Flash-Lite AI Studio, Claude 4.5 Sonnet Vertex, Claude Sonnet 4.6 (Non-reasoning), Gemini 3.5 Flash (minimal) AI Studio, Claude Opus 4.5 Vertex, GLM-4.7, Kimi K2 Thinking Vertex, Claude 4.5 Haiku Vertex, Gemma 4 31B (AI Studio), Claude 4.5 Sonnet Vertex, MiniMax-M2 Vertex, Gemini 3 Flash (AI Studio), GLM-4.7, Gemini 2.5 Pro Vertex, Gemini 2.5 Pro (AI Studio), Gemma 4 26B A4B AI Studio, Gemini 3.1 Flash-Lite (AI Studio), gpt-oss-120b (high) Vertex, Claude 4.5 Haiku Vertex, DeepSeek V3.1 Vertex, DeepSeek V3.1 Vertex, DeepSeek R1 0528 Vertex, Gemini 2.5 Flash (AI Studio), Gemini 2.5 Flash (Vertex), Qwen3 235B 2507 Vertex, Qwen3 Coder 480B Vertex, Qwen3 Next 80B A3B Vertex, gpt-oss-120b (low) Vertex, gpt-oss-20b (high) Vertex, gpt-oss-20b (low) Vertex, Gemini 2.5 Flash (Vertex), Gemini 2.5 Flash (AI Studio), Qwen3 Next 80B A3B Vertex, Gemini 2.5 Flash-Lite (AI Studio), Llama 4 Scout Vertex, Llama 3.3 70B Vertex, and Gemini 2.5 Flash-Lite (AI Studio).

The most intelligent model available on Google is Claude Fable 5 (with fallback) with an Intelligence Index score of 60.

The fastest model on Google by output speed is Gemini 3.5 Flash-Lite AI Studio at 462.6 tokens per second.

The model with the lowest time to first answer token on Google is Gemini 2.5 Flash-Lite (AI Studio) at 0.33s. Lower latency means faster initial response time.

The most affordable model on Google by blended price is Gemini 2.5 Flash-Lite (AI Studio) at $0.07 per 1M tokens (7:2:1 cache hit/input/output ratio).

Prices on Google vary up to 115x across models, from $0.07 per 1M tokens for Gemini 2.5 Flash-Lite (AI Studio) to $7.70 per 1M tokens for Claude Fable 5 (with fallback).

Yes, Google offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.

44 of 52 models on Google support JSON mode for structured output.

Yes, all 52 models on Google support function calling (tool use).

Yes, Google offers 35 reasoning models: Claude Fable 5 (with fallback), Claude Opus 4.8 (max), Claude Opus 4.7 (max), Claude Sonnet 5 (max), Gemini 3.5 Flash AI Studio, Gemini 3.6 Flash AI Studio, Claude Sonnet 4.6 (max), Gemini 3.1 Pro Preview (AI Studio), Gemini 3.1 Pro Preview (Vertex), Gemini 3.5 Flash (medium), Claude Opus 4.6 (max), Claude Opus 4.5 Vertex, GLM-5, Gemini 3 Flash (AI Studio), Gemini 3.5 Flash-Lite AI Studio, Claude 4.5 Sonnet Vertex, GLM-4.7, Kimi K2 Thinking Vertex, Claude 4.5 Haiku Vertex, Gemma 4 31B (AI Studio), MiniMax-M2 Vertex, Gemini 2.5 Pro Vertex, Gemini 2.5 Pro (AI Studio), Gemma 4 26B A4B AI Studio, Gemini 3.1 Flash-Lite (AI Studio), gpt-oss-120b (high) Vertex, DeepSeek V3.1 Vertex, DeepSeek R1 0528 Vertex, Gemini 2.5 Flash (AI Studio), Gemini 2.5 Flash (Vertex), Qwen3 Next 80B A3B Vertex, gpt-oss-120b (low) Vertex, gpt-oss-20b (high) Vertex, gpt-oss-20b (low) Vertex, and Gemini 2.5 Flash-Lite (AI Studio). Reasoning models use extended thinking to work through complex problems before providing an answer.

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 Google, 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.

Google: Models Intelligence, Performance & Price

Most Intelligent

Fastest

Lowest Price

Intelligence