AI models with web search: 28 with a live view of the web

Prices from official vendor price lists · October 2026
28 models that query the live web while answering and cite what they found. Output prices run from $0.28 to $15 per 1M tokens across the 25 models with a paid per-token price; 12 have a free tier.
Web search grounding
Services found: 28
#AI modelCapabilitiesAPI $/1M (in / out)Context windowPrice / moOutput $/1MOverviewVisit site
21

Nano Banana (Gemini 2.5 Flash Image)

Google
VisionImage generationWeb search grounding
$0.3 / $2.532.8K tokensAPI only2.5View
22

Nano Banana Pro (Gemini 3 Pro Image)

Google
VisionAdjustable thinkingImage generation
$2 / $12131.1K tokensAPI only12View
23

Sonar Pro Search

Perplexity
Text & chatVisionReasoning
$3 / $15200K tokensAPI only15View
24

Qwen3.8 Omni Flash

Qwen
Text & chatVisionAudio
1M tokensAPI onlyView
25

MiMo-V2.6-Pro

Xiaomi
Text & chatVisionAudio
$0.435 / $0.871M tokensAPI only0.87View
26

MiMo-V2.6-Flash

Xiaomi
Text & chatVisionAudio
$0.14 / $0.281M tokensAPI only0.28View
27

MiMo-V2.6-Pro-UltraSpeed

Xiaomi
Text & chatVisionAudio
$4.35 / $8.71M tokensAPI only8.7View
28

Fugu Max

Sakana
Text & chatReasoningAdjustable thinking
$2 / $6API only6View

Browse by need

Reasoning models169 modelsCoding assistants17 modelsVision and images156 modelsLong context 1M+82 modelsFree tier available80 modelsOpen weights88 models

Twenty-eight models can query the live web while composing an answer and cite what they found, rather than relying only on training data. Grok 4.6, Gemini 3.8 Flash and the Gemini 3.7 family lead here, with output from $1.50 to $15 per million tokens.

Search is metered separately from tokens: Google charges per 1,000 grounded queries after a monthly free allowance, xAI prices web and X search per thousand calls, and one user question can trigger several searches. Grounding is right for anything time-sensitive - prices, news, availability, regulations - and overhead for stable knowledge. Citation is not verification.

Metered separatelySearch is billed per query on top of tokens, and one question can trigger several.
Freshness, not truthGrounding fixes staleness; it does not fix misreading a source.
Free allowances existGoogle includes a monthly grant before per-query charges begin.

Models with this capability

The 28 models below all support this capability. Prices are input and output per 1M tokens from the vendor official price list; for dirhams multiply by 3.6725. Tap a card to open the model page.

GPT-6.1 Sol
#1
OpenAI
in $2out $101.05M tokens
GPT-6 Sol
#2
OpenAI
in $2out $101.05M tokens
GPT-6 Luna
#3
OpenAI
in $0.1out $0.51.05M tokens
Grok 4.6
#4
xAI
in $2out $6500K tokens
Gemini 3.8 Flash
#5
Google DeepMind
in $0.75out $3.751.05M tokens
Gemini 3.7 Flash
#6
Google DeepMind
in $0.75out $3.751.05M tokens
Grok 4.5
#7
xAI
in $2out $6500K tokens
Gemini 3.6 Flash
#8
Google DeepMind
in $0.75out $3.751.05M tokens

How to choose

  • Restrict grounding to queries that genuinely need fresh data.
  • Log the queries the model issues — they explain most cost surprises.
  • Check whether citations are returned in a structured form you can display.
  • Verify anything regulatory or financial against the primary source.

Common mistakes

Grounding every requestStable knowledge does not need a search, and each one costs.
Treating citations as verificationThe model can cite a page that does not support the claim.

Who this suits

News and financeAnything where yesterday is already wrong.
Travel and commercePrices and availability that change hourly.
ResearchCurrent literature and regulatory updates.

Related filters

Agentic workflows5 models built for autonomous multi-step work
Tool calling166 models that call your functions
Reasoning169 models that think before answering
Frontier tier41 flagship models, the best each vendor sells
All 239 modelsThe full rating with filters and prices

Frequently asked questions

Is search included in the token price?

No. Google charges per 1,000 grounded queries after a free monthly allowance, xAI prices web and X search per thousand calls.

How many searches does one question trigger?

Often several. Budget by query, not by user message.

Does citation mean the answer is verified?

No. Models still misread sources. Citations make checking possible, not unnecessary.

How many searches does one answer use?

Often two to five on a complex question. Budget per query, not per message.

Can I restrict the sources?

Some vendors allow domain filtering; most do not expose it yet.

Frequently searched

By capabilityReasoningCodingVisionLong context 1M+Free tierOpen weightsLow latencyRealtime voiceImage generation
By vendorAnthropic (13)OpenAI (30)Google DeepMind (17)xAI (8)Meta (6)DeepSeek (9)Mistral AI (11)Alibaba (Qwen) (34)Z.ai (12)Moonshot AI (5)
Popular modelsClaude Fable 5.1GPT-6 AstraClaude Opus 5Claude Fable 5GPT-5.6 SolGLM 5.3Grok 4.6Kimi K3GPT-5.6 Terra
By deliveryBatch APIPrompt cachingPriority tierFrontier tierProprietary APITool callingStructured JSONWeb searchMultilingual

Ratings 0

Avg. score 0

Rate this page

Help us improve