how does grok api cost for grok 3 compare with gemeni and sonnet Give me a table of…
- User
how does grok api cost for grok 3 compare with gemeni and sonnet
Give me a table of costs per model
- Assistant
Okay, let's compare the API costs for the latest generally available versions of Grok, Google Gemini, and Anthropic's Sonnet models based on the search results.
It's important to note that while you asked for Grok-3, the search results indicate that the currently available API pricing is for Grok-2 [18, 8, 2]. Grok-3 access seems primarily tied to X subscription plans like Premium+ or SuperGrok rather than direct API token costs [21, 25]. The most comparable models from Google and Anthropic currently are Gemini 1.5 Pro and Claude 3.7 Sonnet, respectively.
Prices are typically quoted per 1 million tokens (MTok). Remember that 1 million tokens roughly equates to about 750,000 words, although this can vary [15, 17].
Here is a table comparing the API costs as of early April 2025:
Model Family Specific Model Input Cost (per 1M tokens) Output Cost (per 1M tokens) Notes xAI Grok Grok-2 (grok-2-1212) $2.00 $10.00 Based on official xAI Docs [18]. Grok-beta is deprecated [18].[[1]] Google Gemini Gemini 1.5 Pro $1.25 (prompts ≤ 128k tokens) $5.00 (prompts ≤ 128k tokens) Based on official Google AI Devs pricing [5]. $2.50 (prompts > 128k tokens) $10.00 (prompts > 128k tokens) Costs vary based on the input prompt size [5].[[2]] Anthropic Claude Claude 3.7 Sonnet $3.00 $15.00 Latest Sonnet model pricing [1, 7, 13].[[3]] Available via Anthropic API, AWS Bedrock, Google Vertex AI [7].[[3]] Summary of Findings:
- Grok-2 API: Offers a relatively low input cost but a mid-range output cost compared to the others listed [18]. Note that Grok access for consumers is often bundled with X Premium subscriptions [2, 12, 22, 25].
- Gemini 1.5 Pro API: Has a variable pricing structure that is generally the most cost-effective, especially for smaller prompts (≤128k tokens) [5]. It becomes competitive with Grok-2 for larger prompts. Google also offers Gemini 1.5 Flash models, which are even cheaper but less capable [5, 19]. Free tiers with rate limits are available [4, 5, 24].[[4]][[5]]
- Claude 3.7 Sonnet API: Positioned as the most intelligent model in the Sonnet line, it has the highest input and output costs among these specific comparable models [1, 7, 13]. Anthropic also offers the cheaper and faster Claude 3.5 Haiku model [13, 14].
Please remember that pricing can change, and providers may offer different tiers, regional variations, or free credits [2, 12, 5]. Always check the official pricing pages for the most current information [5, 13, 18].
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.