# AKI AI Pricing Sources Index™

**Citation Object**: `aki:pricing-sources:top100` | **Edition**: 2026.4.0 | **Updated**: 2026-10-11

> We don't rank AI prices. We rank the websites that report them.

## Independent Trackers (Rankers of the Rankers™)

| Rank | Platform | Overall Trust | Freshness | Accuracy | Coverage | Status |
| :--- | :--- | :--- | :--- | :--- | :--- | :---
| #1 | [ArtificialWatch](https://artificialwatch.ai) | **92.7** / 100 | 96% | 94% | 91% | Independently Audited |
| #2 | [LLMRates.ai](https://llmrates.ai) | **91.1** / 100 | 94% | 91% | 93% | Verified Fresh |
| #3 | [BenchLM Cost Matrix](https://benchlm.org/costs) | **90.2** / 100 | 92% | 93% | 86% | Verified Fresh |
| #4 | [Tokenando API Index](https://tokenando.com) | **89.4** / 100 | 93% | 90% | 89% | Verified Fresh |
| #5 | [AI//COST Monitor](https://aicost.io) | **88.3** / 100 | 91% | 89% | 84% | Discrepancy Flagged |
| #6 | [CloudPrice LLM Feed](https://cloudprice.net/llm) | **87.6** / 100 | 90% | 87% | 95% | Documented |
| #7 | [StackSpend Index](https://stackspend.com) | **85.1** / 100 | 88% | 88% | 82% | Stale — Needs Update |
| #8 | [LLM PriceCheck](https://llmpricecheck.com) | **85.9** / 100 | 89% | 86% | 87% | Documented |
| #9 | [ModelTokens.org](https://modeltokens.org) | **85.1** / 100 | 87% | 85% | 88% | Documented |
| #10 | [PricingMatrix AI](https://pricingmatrix.ai) | **83.0** / 100 | 86% | 84% | 85% | Limited Transparency |
| #11 | [AI Inference Cost Index](https://aiinferencecost.com) | **85.3** / 100 | 88% | 87% | 80% | Documented |
| #12 | [TokenCostHub](https://tokencosthub.com) | **81.6** / 100 | 85% | 82% | 83% | Limited Transparency |
| #13 | [InferenceRadar](https://inferenceradar.io) | **88.7** / 100 | 91% | 90% | 82% | Verified Fresh |
| #14 | [API Price Tracker](https://apipricetracker.org) | **79.7** / 100 | 84% | 81% | 79% | Stale — Needs Update |

## Official Direct Providers

| Rank | Provider | Official API URL | Freshness | Trust Score | Status |
| :--- | :--- | :--- | :--- | :--- | :---
| #1 | **OpenAI Official Pricing** | [API Pricing](https://openai.com/api/pricing) | 98% | **96.4** / 100 | Verified Fresh |
| #2 | **Anthropic Official Pricing** | [API Pricing](https://anthropic.com/pricing) | 97% | **95.1** / 100 | Verified Fresh |
| #3 | **Google Cloud Vertex / AI Studio** | [API Pricing](https://cloud.google.com/vertex-ai/pricing) | 95% | **94.3** / 100 | Verified Fresh |
| #4 | **DeepSeek Open Platform** | [API Pricing](https://platform.deepseek.com/pricing) | 94% | **92.8** / 100 | Verified Fresh |
| #5 | **xAI Developer Console** | [API Pricing](https://x.ai/api) | 93% | **90.6** / 100 | Documented |
| #6 | **Together AI Model Inference** | [API Pricing](https://www.together.ai/pricing) | 94% | **93.5** / 100 | Verified Fresh |
| #7 | **Mistral AI La Plateforme** | [API Pricing](https://mistral.ai/technology/#pricing) | 92% | **91.9** / 100 | Verified Fresh |
| #8 | **Cohere Platform Pricing** | [API Pricing](https://cohere.com/pricing) | 91% | **91.2** / 100 | Verified Fresh |
| #9 | **Amazon Bedrock API Pricing** | [API Pricing](https://aws.amazon.com/bedrock/pricing/) | 90% | **91.8** / 100 | Verified Fresh |
| #10 | **Azure OpenAI Service** | [API Pricing](https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/) | 91% | **92.2** / 100 | Verified Fresh |
| #11 | **Fireworks AI Fast Inference** | [API Pricing](https://fireworks.ai/pricing) | 95% | **93.5** / 100 | Verified Fresh |
| #12 | **Groq LPU Cloud Pricing** | [API Pricing](https://groq.com/pricing) | 96% | **93.1** / 100 | Verified Fresh |
| #13 | **Replicate Model Hub** | [API Pricing](https://replicate.com/pricing) | 90% | **90.8** / 100 | Documented |
| #14 | **Anyscale Endpoints** | [API Pricing](https://www.anyscale.com/endpoints) | 89% | **89.8** / 100 | Documented |
| #15 | **01.AI Yi Developer Platform** | [API Pricing](https://platform.lingyiwanwu.com/) | 88% | **87.8** / 100 | Documented |
| #16 | **Cerebras CS-3 Fast Inference** | [API Pricing](https://cerebras.ai/pricing) | 94% | **90.8** / 100 | Verified Fresh |

## Discrepancy Radar

| Model & Tier | Official (GBP) | Reported (GBP) | Reporting Source | Variance | Root Cause |
| :--- | :--- | :--- | :--- | :--- | :---
| **Claude 3.5 Sonnet (Input/1M)** | £2.35 | £2.85 | AI//COST Monitor | **+21.3%** | Currency Conversion Drift |
| **GPT-4o Mini (Output/1M)** | £0.47 | £0.60 | CloudPrice LLM Feed | **+27.7%** | Stale Price Cache |
| **DeepSeek-V2.5 (Input/1M)** | £0.11 | £0.14 | StackSpend Index | **+27.3%** | Token Unit Mismatch (1k vs 1M) |
| **Gemini 1.5 Pro (Cached Prompt/1M)** | £0.68 | £0.95 | PricingMatrix AI | **+39.7%** | Unannounced Batch vs Realtime Tier |
| **Llama 3.1 405B (Output/1M)** | £3.90 | £4.80 | TokenCostHub | **+23.1%** | Currency Conversion Drift |
| **Mistral Large 2 (Prompt/1M)** | £1.55 | £1.95 | API Price Tracker | **+25.8%** | Stale Price Cache |
| **Grok 2 Beta (Input/1M)** | £1.56 | £2.10 | LLM PriceCheck | **+34.6%** | Token Unit Mismatch (1k vs 1M) |
| **Command R+ (Input/1M)** | £1.95 | £2.45 | AI//COST Monitor | **+25.6%** | Currency Conversion Drift |
| **Qwen 2.5 72B (Hosted/1M)** | £0.28 | £0.38 | CloudPrice LLM Feed | **+35.7%** | Stale Price Cache |
| **Claude 3 Opus (Prompt/1M)** | £11.75 | £14.20 | StackSpend Index | **+20.9%** | Currency Conversion Drift |
| **GPT-4o Realtime Audio (Input/1M)** | £78.50 | £102.00 | PricingMatrix AI | **+29.9%** | Unannounced Batch vs Realtime Tier |
| **Fireworks Mixtral 8x22B (Output/1M)** | £0.70 | £0.95 | ModelTokens.org | **+35.7%** | Token Unit Mismatch (1k vs 1M) |
| **Groq Llama 3 8B (Prompt/1M)** | £0.04 | £0.06 | API Price Tracker | **+50.0%** | Stale Price Cache |
| **DeepSeek-Coder-V2 (Output/1M)** | £0.22 | £0.28 | TokenCostHub | **+27.3%** | Currency Conversion Drift |

## Price Velocity Tracker

| Model | Provider | Shift (GBP) | Change % | Fastest Source | Detection Lag |
| :--- | :--- | :--- | :--- | :--- | :---
| **Gemini 1.5 Flash (Prompt/1M)** | Google DeepMind | £0.28 → £0.06 | **-78.6%** | ArtificialWatch | 1.2 hrs |
| **GPT-4o (Output/1M)** | OpenAI | £11.75 → £7.85 | **-33.2%** | LLMRates.ai | 2.4 hrs |
| **Llama 3.1 70B (Hosted/1M)** | Together AI | £0.70 → £0.42 | **-40.0%** | BenchLM Cost Matrix | 3.1 hrs |
| **DeepSeek-V2.5 (Output/1M)** | DeepSeek AI | £0.22 → £0.11 | **-50.0%** | Tokenando API Index | 0.8 hrs |
| **Claude 3.5 Sonnet (Prompt Cache)** | Anthropic | £2.35 → £0.24 | **-89.8%** | ArtificialWatch | 1.5 hrs |
| **Mistral Small (Prompt/1M)** | Mistral AI | £0.78 → £0.16 | **-79.5%** | LLMRates.ai | 2.1 hrs |
| **Fireworks Qwen 2.5 72B (Prompt/1M)** | Fireworks AI | £0.70 → £0.28 | **-60.0%** | InferenceRadar | 1.9 hrs |
| **Cerebras Llama 3.1 8B (Speed/1M)** | Cerebras | £0.16 → £0.08 | **-50.0%** | BenchLM Cost Matrix | 2.8 hrs |
| **Groq Llama 3.1 70B (Output/1M)** | Groq | £0.62 → £0.46 | **-25.8%** | Tokenando API Index | 3.5 hrs |
| **Amazon Bedrock Nova Micro (Input/1M)** | AWS | £0.12 → £0.03 | **-75.0%** | ArtificialWatch | 1.4 hrs |
| **Azure GPT-4o Batch API (Prompt/1M)** | Microsoft Azure | £1.95 → £0.98 | **-49.7%** | LLMRates.ai | 2.0 hrs |
| **Cohere Command R (Prompt/1M)** | Cohere | £0.39 → £0.12 | **-69.2%** | InferenceRadar | 2.6 hrs |