Guides
Pick the right model and keep the bill down. Every guide is grounded in current API prices.
Best LLM for…
Best LLM for Coding (2026)
Code generation rewards strong reasoning and a large context window so the model can see your whole file or repo. These are the mo…
Cheapest LLM API That Is Still Good (2026)
When you are processing huge volumes — classification, extraction, summarisation — price per token is everything. These are the lo…
Best LLM for Long Context (2026)
Feeding whole books, codebases or document sets to a model needs a large context window — and the cost scales with every token you…
Best LLM for Reasoning & Math (2026)
Hard math, planning and multi-step logic are where reasoning models earn their premium. They "think" before answering, which means…
Model price comparisons
GPT-6 Astra vs Claude Fable 5.1
Compare equal-token budgets, cache writes and reads, and Astra’s 272K long-context price boundary.
Haiku 5.5 vs Gemini 3.8 Flash
Compare prompt bands, promotional prices, cache storage and processing modes.
Haiku 5.5 vs Opus 5.5
Choose a model by workload, compare Standard and Fast costs, and open both models in the calculator with the same token budget.
Sol 6.1 vs Sol 6 vs Astra
Compare official token prices and estimate cache reads, cache writes, long prompts and processing modes for your workload.
API updates
Grok Build pricing
Free access, account subscriptions and separate API token costs, with the Grok Code Fast alias explained.
Gemini 3.1 Flash-Lite pricing
Standard, Batch, Flex and Priority rates, audio costs, cache storage and a reproducible $40 budget.
Microsoft Decision-1 pricing
Input-only token costs, Foundry and OpenRouter access, and reproducible routing budgets.
Claude Sonnet 5.5 pricing
Standard, cache-write and Batch budgets across the 1M context window.
Perplexity Decider API pricing
Input-only billing for v1.1 and v1, question and image costs, and routing examples.
Claude Haiku 5.5 API pricing
Check the 100K prompt boundary, cache writes, Batch rates and migration costs with reproducible examples.
OpenAI Decisions API pricing
GPT-6 Luna's decision endpoint: input-only billing, the 10× speed claim and billing exceptions. Includes links to the new usage tiers.
Articles
GPT-5.6 Pricing: Sol, Terra & Luna API Cost Compared
GPT-5.6 API pricing — Sol $5/$30, Terra $2.50/$15, Luna $1/$6 per 1M tokens. How the three tiers compare, worked cost examples, prompt caching, and how they stack up against GPT-5.5, Claude and Gemini.
GPT-5.6 vs GPT-5.5: Pricing, Tiers & Whether to Upgrade
GPT-5.6 (Sol, Terra, Luna) vs GPT-5.5 — what changed, the price differences, and when each is the cheaper choice. Plus how GPT-5.6 compares to Claude Fable 5 and Gemini 2.5 Pro.
How to Estimate LLM API Costs Before You Build
A practical walkthrough for estimating your LLM API bill from tokens and request volume — with worked examples and the levers that actually move the number.
OpenAI vs Anthropic vs Google: API Pricing Compared
How the big three LLM providers price their APIs, where each is cheapest, and how to choose between GPT-5, Claude and Gemini on cost.
Input vs Output Tokens: Why Your LLM Bill Is Higher Than You Think
Output tokens cost far more than input tokens on almost every LLM API. Here is why, and how to design prompts and limits around it.