Pricing Changelog

Every price change, audit, and update — dated and documented.

2026-07-04

Added 6 new models — now 26 models from 12 providers

Market audit identified gaps vs. mainstream LLM landscape. Added Grok 4.3, Cohere Command A+, Qwen 3.7 Max, Amazon Nova Premier, Llama 4 Scout, and Mistral Small 3.2. Comparison pages grew from 190 to 325.

+xAI Grok 4.3 — $1.25/$2.50, 1M context, video input, best US frontier output price
+Cohere Command A+ — free during preview, 128 experts MoE, Apache 2.0 open weights
+Alibaba Qwen 3.7 Max — $2.50/$7.50, MMLU-Pro 89.6, Alibaba's true flagship (replaces Qwen 3 Coder as top offering)
+Amazon Nova Premier — $2.50/$12.50, AWS Bedrock native, 1M context
+Meta Llama 4 Scout — $0.10/$0.30, 10M context window (largest of any model listed)
+Mistral Small 3.2 — $0.075/$0.20, EU data residency, cheapest multimodal model
+Benchmark types updated to support null for models without published scores

2026-06-30

Site launched

AI Model Cost Calculator launched with 19 models across 9 providers. All prices sourced from official API pricing pages.

+19 models from Anthropic, OpenAI, Google, DeepSeek, Alibaba, Moonshot AI, MiniMax, Mistral, Meta
+Support for prompt caching, batch processing, and long context pricing
+8 quick presets for common usage scenarios

2026-06-29

Initial pricing data compiled

Verified all model prices against official provider pricing pages. Last full audit before launch.

+Anthropic: Claude Opus 4.8, Sonnet 4.6, Haiku 4.5, Fable 5 verified
+OpenAI: GPT-5.4, GPT-5 Sol, GPT-5 Mini, GPT-5 Micro, o4-mini verified
+Google: Gemini 3.1 Pro, Gemini 3.1 Flash, Gemini 3.1 Ultra verified
+DeepSeek: DeepSeek-V4 Pro, DeepSeek-V4 Flash verified
+Others: Qwen3-Plus, Kimi K2, MiniMax-M1, Mistral Large 3, Llama 4 Maverick verified