Thangavel PrasanthTP

Gemini comparison

5 min read

Model comparison

  • We can read about the available models and their capabilities on the model overview page
  • The table below is extracted from Kaggle notebook as part of Google-5-Day-Gen-AI-Intensive-Course. As of writing this note (on 23-May-2025),
    • It had 56 models under the hood.
    • Some of the models support an
      • input_token_limit of 2000000
      • output_token_limitof 65536
    • Different actions supported are
      • createTunedTextModel
      • generateContent
      • generateMessage
      • predict
      • createTunedModel
      • countTokens
      • createCachedContent
      • bidiGenerateContent
      • embedText
      • embedContent
      • countTextTokens
      • countMessageTokens
      • generateAnswer
namedisplay_namedescriptionversioninput_token_limitoutput_token_limitsupported_actions
models/chat-bison-001PaLM 2 Chat (Legacy)A legacy text-only model optimized for chat conversations00140961024[‘generateMessage’, ‘countMessageTokens’]
models/text-bison-001PaLM 2 (Legacy)A legacy model that understands text and generates text as an output00181961024[‘generateText’, ‘countTextTokens’, ‘createTunedTextModel’]
models/embedding-gecko-001Embedding GeckoObtain a distributed representation of a text.00110241[‘embedText’, ‘countTextTokens’]
models/gemini-1.0-pro-vision-latestGemini 1.0 Pro VisionThe original Gemini 1.0 Pro Vision model version which was optimized for image understanding. Gemini 1.0 Pro Vision was deprecated on July 12, 2024. Move to a newer Gemini version.001122884096[‘generateContent’, ‘countTokens’]
models/gemini-pro-visionGemini 1.0 Pro VisionThe original Gemini 1.0 Pro Vision model version which was optimized for image understanding. Gemini 1.0 Pro Vision was deprecated on July 12, 2024. Move to a newer Gemini version.001122884096[‘generateContent’, ‘countTokens’]
models/gemini-1.5-pro-latestGemini 1.5 Pro LatestAlias that points to the most recent production (non-experimental) release of Gemini 1.5 Pro, our mid-size multimodal model that supports up to 2 million tokens.00120000008192[‘generateContent’, ‘countTokens’]
models/gemini-1.5-pro-001Gemini 1.5 Pro 001Stable version of Gemini 1.5 Pro, our mid-size multimodal model that supports up to 2 million tokens, released in May of 2024.00120000008192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-1.5-pro-002Gemini 1.5 Pro 002Stable version of Gemini 1.5 Pro, our mid-size multimodal model that supports up to 2 million tokens, released in September of 2024.00220000008192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-1.5-proGemini 1.5 ProStable version of Gemini 1.5 Pro, our mid-size multimodal model that supports up to 2 million tokens, released in May of 2024.00120000008192[‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-latestGemini 1.5 Flash LatestAlias that points to the most recent production (non-experimental) release of Gemini 1.5 Flash, our fast and versatile multimodal model for scaling across diverse tasks.00110000008192[‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-001Gemini 1.5 Flash 001Stable version of Gemini 1.5 Flash, our fast and versatile multimodal model for scaling across diverse tasks, released in May of 2024.00110000008192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-1.5-flash-001-tuningGemini 1.5 Flash 001 TuningVersion of Gemini 1.5 Flash that supports tuning, our fast and versatile multimodal model for scaling across diverse tasks, released in May of 2024.001163848192[‘generateContent’, ‘countTokens’, ‘createTunedModel’]
models/gemini-1.5-flashGemini 1.5 FlashAlias that points to the most recent stable version of Gemini 1.5 Flash, our fast and versatile multimodal model for scaling across diverse tasks.00110000008192[‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-002Gemini 1.5 Flash 002Stable version of Gemini 1.5 Flash, our fast and versatile multimodal model for scaling across diverse tasks, released in September of 2024.00210000008192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-1.5-flash-8bGemini 1.5 Flash-8BStable version of Gemini 1.5 Flash-8B, our smallest and most cost effective Flash model, released in October of 2024.00110000008192[‘createCachedContent’, ‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-8b-001Gemini 1.5 Flash-8B 001Stable version of Gemini 1.5 Flash-8B, our smallest and most cost effective Flash model, released in October of 2024.00110000008192[‘createCachedContent’, ‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-8b-latestGemini 1.5 Flash-8B LatestAlias that points to the most recent production (non-experimental) release of Gemini 1.5 Flash-8B, our smallest and most cost effective Flash model, released in October of 2024.00110000008192[‘createCachedContent’, ‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-8b-exp-0827Gemini 1.5 Flash 8B Experimental 0827Experimental release (August 27th, 2024) of Gemini 1.5 Flash-8B, our smallest and most cost effective Flash model. Replaced by Gemini-1.5-flash-8b-001 (stable).00110000008192[‘generateContent’, ‘countTokens’]
models/gemini-1.5-flash-8b-exp-0924Gemini 1.5 Flash 8B Experimental 0924Experimental release (September 24th, 2024) of Gemini 1.5 Flash-8B, our smallest and most cost effective Flash model. Replaced by Gemini-1.5-flash-8b-001 (stable).00110000008192[‘generateContent’, ‘countTokens’]
models/gemini-2.5-pro-exp-03-25Gemini 2.5 Pro Experimental 03-25Experimental release (March 25th, 2025) of Gemini 2.5 Pro2.5-exp-03-25104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.5-pro-preview-03-25Gemini 2.5 Pro Preview 03-25Gemini 2.5 Pro Preview 03-252.5-preview-03-25104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.5-flash-preview-04-17Gemini 2.5 Flash Preview 04-17Preview release (April 17th, 2025) of Gemini 2.5 Flash2.5-preview-04-17104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-expGemini 2.0 Flash ExperimentalGemini 2.0 Flash Experimental2.010485768192[‘generateContent’, ‘countTokens’, ‘bidiGenerateContent’]
models/gemini-2.0-flashGemini 2.0 FlashGemini 2.0 Flash2.010485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-001Gemini 2.0 Flash 001Stable version of Gemini 2.0 Flash, our fast and versatile multimodal model for scaling across diverse tasks, released in January of 2025.2.010485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-lite-001Gemini 2.0 Flash-Lite 001Stable version of Gemini 2.0 Flash Lite2.010485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-liteGemini 2.0 Flash-LiteGemini 2.0 Flash-Lite2.010485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-lite-preview-02-05Gemini 2.0 Flash-Lite Preview 02-05Preview release (February 5th, 2025) of Gemini 2.0 Flash Litepreview-02-0510485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-lite-previewGemini 2.0 Flash-Lite PreviewPreview release (February 5th, 2025) of Gemini 2.0 Flash Litepreview-02-0510485768192[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-pro-expGemini 2.0 Pro ExperimentalExperimental release (March 25th, 2025) of Gemini 2.5 Pro2.5-exp-03-25104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-pro-exp-02-05Gemini 2.0 Pro Experimental 02-05Experimental release (March 25th, 2025) of Gemini 2.5 Pro2.5-exp-03-25104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-exp-1206Gemini Experimental 1206Experimental release (March 25th, 2025) of Gemini 2.5 Pro2.5-exp-03-25104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-thinking-exp-01-21Gemini 2.5 Flash Preview 04-17Preview release (April 17th, 2025) of Gemini 2.5 Flash2.5-preview-04-17104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-thinking-expGemini 2.5 Flash Preview 04-17Preview release (April 17th, 2025) of Gemini 2.5 Flash2.5-preview-04-17104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/gemini-2.0-flash-thinking-exp-1219Gemini 2.5 Flash Preview 04-17Preview release (April 17th, 2025) of Gemini 2.5 Flash2.5-preview-04-17104857665536[‘generateContent’, ‘countTokens’, ‘createCachedContent’]
models/learnlm-1.5-pro-experimentalLearnLM 1.5 Pro ExperimentalAlias that points to the most recent stable version of Gemini 1.5 Pro, our mid-size multimodal model that supports up to 2 million tokens.001327678192[‘generateContent’, ‘countTokens’]
models/learnlm-2.0-flash-experimentalLearnLM 2.0 Flash ExperimentalLearnLM 2.0 Flash Experimental2.0104857632768[‘generateContent’, ‘countTokens’]
models/gemma-3-1b-itGemma 3 1Bnan001327688192[‘generateContent’, ‘countTokens’]
models/gemma-3-4b-itGemma 3 4Bnan001327688192[‘generateContent’, ‘countTokens’]
models/gemma-3-12b-itGemma 3 12Bnan001327688192[‘generateContent’, ‘countTokens’]
models/gemma-3-27b-itGemma 3 27Bnan0011310728192[‘generateContent’, ‘countTokens’]
models/embedding-001Embedding 001Obtain a distributed representation of a text.00120481[‘embedContent’]
models/text-embedding-004Text Embedding 004Obtain a distributed representation of a text.00420481[‘embedContent’]
models/gemini-embedding-exp-03-07Gemini Embedding Experimental 03-07Obtain a distributed representation of a text.exp-03-0781921[‘embedContent’, ‘countTextTokens’]
models/gemini-embedding-expGemini Embedding ExperimentalObtain a distributed representation of a text.exp-03-0781921[‘embedContent’, ‘countTextTokens’]
models/aqaModel that performs Attributed Question Answering.Model trained to return answers to questions that are grounded in provided sources, along with estimating answerable probability.00171681024[‘generateAnswer’]
models/imagen-3.0-generate-002Imagen 3.0 002 modelVertex served Imagen 3.0 002 model0024808192[‘predict’]
models/gemini-2.0-flash-live-001Gemini 2.0 Flash 001Gemini 2.0 Flash 0010011310728192[‘bidiGenerateContent’, ‘countTokens’]

Rate limits

ModelRPMTPMRPD
Gemini 2.5 Flash Preview 04-1710250,000500
Gemini 2.5 Pro Experimental5250,00025
Gemini 2.5 Pro Preview———
Gemini 2.0 Flash151,000,0001,500
Gemini 2.0 Flash Experimental (including image generation)101,000,0001,500
Gemini 2.0 Flash-Lite301,000,0001,500
Gemini 1.5 Flash151,000,0001,500
Gemini 1.5 Flash-8B151,000,0001,500
…………

Gemini 2.5 Flash Preview

Google’s first hybrid reasoning model which supports a 1M token context and has thinking budgets.

Free TierPaid Tier, per 1M tokens in USD
Input priceFree of charge1.00 (audio)
Output priceFree of chargeNon-thinking: 3.50
Context caching priceNot available0.25 (audio)
$1.00 / 1,000,000 tokens per hour
Grounding with Google SearchFree of charge, up to 500 RPD1,500 RPD (free), then $35 / 1,000 requests
Text-to-speech
(gemini-2.5-flash-preview-tts)		Free of charge10.00 (Output)
Used to improve our productsYesNo

Gemini API Pricing

https://ai.google.dev/gemini-api/docs/pricing

Graph View

Backlinks