Corrections
When a published article turns out to be wrong, or a fact in it has changed, the article is updated and the change is recorded here. Articles are never quietly edited.
Found an error? Write to ops@3ammarketer.com with the article and what is wrong. Substantive errors are corrected and logged here.
Log
Mistral API Pricing: 2026 Model Token Costs Guide
Mistral Medium 3.5, Mistral Large 3 and mistral-large-2512 are no longer on Mistral's pricing pages. The price range and the cost comparison now use Mistral Large 4, the most expensive current model ($0.70 input, $2.10 output per million tokens).
Our pricing tracker records Medium 3.5 and Large 3 as removed from Mistral's pricing pages; Mistral's model page lists Large 4 at $0.70 per million input tokens and $2.10 per million output tokens.
Mistral Medium 3.5, Mistral Large 3 and mistral-large-2512 are no longer on Mistral's pricing pages. The price range and the cost comparison now use Mistral Large 4, the most expensive current model ($0.70 input, $2.10 output per million tokens).
Our pricing tracker records Medium 3.5 and Large 3 as removed from Mistral's pricing pages; Mistral's model page lists Large 4 at $0.70 per million input tokens and $2.10 per million output tokens.
LM Studio vs Ollama: License Terms and Published Speeds
Ollama's decision model support is the AutoTrust decision models (autotrust-jev), run with ollama run --ask. The earlier text described a /v1/systemone endpoint and question types that the release notes do not mention.
Ollama's release notes describe AutoTrust decision models that "read text and return choices, yes or no probabilities, and ratings"; no /v1/systemone endpoint is listed.
Ollama's decision model support is the AutoTrust decision models (autotrust-jev), run with ollama run --ask. The earlier text described a /v1/systemone endpoint and question types that the release notes do not mention.
Ollama's release notes describe AutoTrust decision models that "read text and return choices, yes or no probabilities, and ratings"; no /v1/systemone endpoint is listed.
Mistral Voxtral 3 and Large 4 Multimodal Release
The price for Mistral Large 4 decreased.
Our pricing tracker shows Mistral Large 4 prices fell as of 4 Nov 2026.
The price for Mistral Large 4 decreased.
Our pricing tracker shows Mistral Large 4 prices fell as of 4 Nov 2026.
SEO Services Pricing: 2026 Rates and Retainers
Updated Coalition Technologies campaign pricing for metropolitan-area and regulated service provider campaigns.
Coalition Technologies now lists metropolitan-area campaigns at an average of $2,000 per month and regulated service provider campaigns at an average of $4,200 per month.
Updated Coalition Technologies campaign pricing for metropolitan-area and regulated service provider campaigns.
Coalition Technologies now lists metropolitan-area campaigns at an average of $2,000 per month and regulated service provider campaigns at an average of $4,200 per month.
Gemini 4 Argon Pricing: $2 Intro Input, 1M Token Output
Added the end date of Gemini 4 Argon's introductory pricing: $2 input and $10 output per million tokens through November 30, 2026, then $4 and $20 from December 1, 2026.
When this entry was published Google had not announced an end date; its Gemini API pricing page now states one.
Added the end date of Gemini 4 Argon's introductory pricing: $2 input and $10 output per million tokens through November 30, 2026, then $4 and $20 from December 1, 2026.
When this entry was published Google had not announced an end date; its Gemini API pricing page now states one.
Claude Prompt Caching: TTL Multipliers and Pricing
Corrected the largest cache read saving from 95 to 97.5 percent and added Claude Mythos 5.1 to the 0.025x cache read rate.
Anthropic prices cache hits on Claude Fable 5.1 and Claude Mythos 5.1 at 0.025x the base input price (12 Oct 2026).
Corrected the largest cache read saving from 95 to 97.5 percent and added Claude Mythos 5.1 to the 0.025x cache read rate.
Anthropic prices cache hits on Claude Fable 5.1 and Claude Mythos 5.1 at 0.025x the base input price (12 Oct 2026).
Claude API Pricing per Token: $0.10 on Haiku to $10 on Fable
Corrected the cached input price for Claude Opus 5.5 from $0.40 to $0.20 per million tokens, and noted that the 10 percent cache read rate applies to most models, not all.
Anthropic prices cache hits on Claude Opus 5.5 at 0.05x the base input price, not the standard 0.1x, and Claude Fable 5.1 at 0.025x.
Corrected the cached input price for Claude Opus 5.5 from $0.40 to $0.20 per million tokens, and noted that the 10 percent cache read rate applies to most models, not all.
Anthropic prices cache hits on Claude Opus 5.5 at 0.05x the base input price, not the standard 0.1x, and Claude Fable 5.1 at 0.025x.