Unbiased: blog and product updates Posts from the Unbiased blog and entries from the changelog, newest first. https://unbiased.ai/feed.xml 2026-10-21T00:00:00Z Unbiasedhttps://unbiased.ai/ https://unbiased.ai/assets/unbiased-icon.png Anthropic API Alternatives in 2026: Prices, Trade-offs, Receipts https://unbiased.ai/blog/anthropic-api-alternatives/ 2026-10-21T00:00:00Z Real Anthropic API alternatives with October 2026 prices: OpenAI's ladder, open weights, gateways, and a blended model, plus the Fable 5 credits change. LiteLLM Alternatives: When Self-Hosting Stops Being Free https://unbiased.ai/blog/litellm-alternatives/ 2026-10-21T00:00:00Z LiteLLM is free until you count ops time. Managed alternatives compared: OpenRouter, Portkey, Requesty, Pareto, with the honest math on when each wins. OpenAI API Alternatives in 2026: Prices, Trade-offs, Receipts https://unbiased.ai/blog/openai-api-alternatives/ 2026-10-21T00:00:00Z Real OpenAI API alternatives with October 2026 prices: Anthropic's ladder, open-weights self-hosting, gateways, and a blended model, with the trade-offs. Opus 4.8 vs Opus 5: Same Price, 77% Different Bills | Unbiased https://unbiased.ai/blog/opus-4-8-vs-opus-5/ 2026-10-21T00:00:00Z Claude Opus 5 and Opus 4.8 share a $5/$25 rate card. On an identical six-task battery, Opus 5 billed 77% more, purely token count. Measured data. Portkey Alternatives: Control Plane, Gateway, or Neither https://unbiased.ai/blog/portkey-alternatives/ 2026-10-21T00:00:00Z Portkey alternatives compared honestly: LiteLLM self-hosted, OpenRouter, Requesty, Pareto. The fit depends on whether the problem is visibility or cost. GPT-6.1 Sol vs GPT-6 Luna: Which Tier, Explained | Unbiased https://unbiased.ai/blog/sol-vs-terra-vs-luna/ 2026-10-21T00:00:00Z GPT-6.1 Sol ($2/$10) vs GPT-6 Luna ($0.10/$0.50 per million tokens): OpenAI's two main API tiers, with GPT-6 Astra above. Which fits which job, with the math. Claude Sonnet 5 Pricing: $2/$10 Is Now Permanent https://unbiased.ai/blog/claude-sonnet-5-pricing/ 2026-10-19T00:00:00Z Sonnet 5 costs $2/$10 per million tokens, and the 50% increase scheduled for Sept 1 was cancelled on Aug 10. Measured bills and what the reversal means. LLM API Pricing: The Complete Guide, With Measured Bills (2026) https://unbiased.ai/blog/llm-api-pricing/ 2026-10-19T00:00:00Z Every major model's October 2026 rates, real metered bills per task, the caching and batch levers, and a calculator for your own traffic. The Sonnet 5 Price Increase That Did Not Happen https://unbiased.ai/blog/the-price-increase-that-did-not-happen/ 2026-10-19T00:00:00Z A 50% increase was scheduled for September 1; Anthropic cancelled it on August 10. What it would have cost, why we ran a countdown clock, and the lesson. Product updates: October 18, 2026 tag:unbiased.ai,2026:changelog/2026-10-18/tokenizer-tax-tool 2026-10-18T00:00:00Z Unbiased changelog entry for October 18, 2026. Pareto 26.10 is out of preview https://unbiased.ai/blog/pareto-26-10/ 2026-10-15T00:00:00Z The preview price is now permanent, the context window is 1M, and subscribers get twice the tokens. Product updates: October 15, 2026 tag:unbiased.ai,2026:changelog/2026-10-15/pareto-26-10-ga 2026-10-15T00:00:00Z Unbiased changelog entry for October 15, 2026. Unbiased Chat is live https://unbiased.ai/blog/unbiased-chat/ 2026-10-09T00:00:00Z Pareto in a chat window. Several models on every question, one answer, $10 a month. All of a Sudden, Elon Cares About Routing https://unbiased.ai/blog/all-of-a-sudden-elon-cares-about-routing/ 2026-10-07T00:00:00Z The “best model for the job” idea just went mainstream. Here’s what you should want from it. Claude Opus 4.8 Pricing: $5/$25, and Why It Still Matters https://unbiased.ai/blog/claude-opus-4-8-pricing/ 2026-10-01T00:00:00Z Opus 4.8 costs $5/$25 per million tokens, the same as Opus 5, and our measured runs show it bills less per task than its successor. The receipts inside. Claude Opus 5 Pricing: Same Rate Card as 4.8, Bigger Bills https://unbiased.ai/blog/claude-opus-5-pricing/ 2026-10-01T00:00:00Z Opus 5 costs $5/$25 per million tokens, unchanged from 4.8 and half of Fable 5. Our measured runs bill 60% more per task. Here is why, with receipts. Fable 5 Alternatives: What Actually Comes Close, With Prices https://unbiased.ai/blog/fable-5-alternatives/ 2026-10-01T00:00:00Z Five real alternatives to Claude Fable 5 with measured costs: Opus 5 at half price, GPT-6.1 Sol, Sonnet 5.5, and a blended model. The trade-offs of each. Fable 5 Pricing Explained: Cost Per Task, vs Opus, Real Bills https://unbiased.ai/blog/fable-5-pricing/ 2026-10-01T00:00:00Z Fable 5 costs $10/$50 per million tokens, exactly 2x Opus. Real measured bills per request, caching and batch discounts, and pricing for your traffic. GPT-6.1 Sol Pricing: $2/$10, Same Rates as Sonnet 5.5 https://unbiased.ai/blog/gpt-6-1-sol-pricing/ 2026-10-01T00:00:00Z GPT-6.1 Sol costs $2 input / $10 output per million tokens, $0.10 cached: Claude Sonnet 5.5's rate card, 60% below Opus 5. The cache math and the market table. GPT-6 Luna Pricing: $0.10/$0.50, a Tenth of Haiku 4.5 https://unbiased.ai/blog/gpt-6-luna-pricing/ 2026-10-01T00:00:00Z GPT-6 Luna costs $0.10 input / $0.50 output per million tokens, a tenth of Claude Haiku 4.5's $1/$5. List-price math beside our measured Claude bills. LLM Gateways: The Complete Guide, With Measured Latency (2026) https://unbiased.ai/blog/llm-gateways/ 2026-10-01T00:00:00Z What an LLM gateway is, the four types, real fees (OpenRouter, Portkey, Requesty, LiteLLM), our measured +68ms hop latency, and how to pick one. Pareto 26.10 launch notes (October 1, 2026) https://unbiased.ai/blog/pareto-26-10-preview/ 2026-10-01T00:00:00Z The October 1 release notes: new token prices and preliminary results on four benchmarks. Now generally available; see the GA post. The Verbosity Index: Tokens Billed for Identical Work | Unbiased https://unbiased.ai/blog/verbosity-index/ 2026-10-01T00:00:00Z Edition 1: Opus 5 writes 48% more tokens than the median model on identical tasks; its same-priced predecessor writes 8% less. Measured monthly. Why Your LLM Bill Does Not Match Your Math https://unbiased.ai/blog/why-your-bill-does-not-match-your-math/ 2026-10-01T00:00:00Z Five measured reasons rate-card arithmetic misses your invoice: verbosity, the tokenizer change, cache misses, retry doubling, and fee stacking. Product updates: October 1, 2026 tag:unbiased.ai,2026:changelog/2026-10-01 2026-10-01T00:00:00Z Unbiased changelog entry for October 1, 2026. Is Your Harness On The Frontier? https://unbiased.ai/blog/is-your-harness-on-the-frontier/ 2026-09-20T00:00:00Z The Unbiased macOS app is now open source. Use your agent to help build the coding tools you’d rather be using. Product updates: September 17, 2026 tag:unbiased.ai,2026:changelog/2026-09-17 2026-09-17T00:00:00Z Unbiased changelog entry for September 17, 2026. Product updates: September 16, 2026 tag:unbiased.ai,2026:changelog/2026-09-16 2026-09-16T00:00:00Z Unbiased changelog entry for September 16, 2026. Product updates: September 15, 2026 tag:unbiased.ai,2026:changelog/2026-09-15 2026-09-15T00:00:00Z Unbiased changelog entry for September 15, 2026. Product updates: September 14, 2026 tag:unbiased.ai,2026:changelog/2026-09-14 2026-09-14T00:00:00Z Unbiased changelog entry for September 14, 2026. Product updates: September 11, 2026 tag:unbiased.ai,2026:changelog/2026-09-11 2026-09-11T00:00:00Z Unbiased changelog entry for September 11, 2026. Product updates: September 10, 2026 tag:unbiased.ai,2026:changelog/2026-09-10 2026-09-10T00:00:00Z Unbiased changelog entry for September 10, 2026. Product updates: September 4, 2026 tag:unbiased.ai,2026:changelog/2026-09-04 2026-09-04T00:00:00Z Unbiased changelog entry for September 4, 2026. Product updates: September 2, 2026 tag:unbiased.ai,2026:changelog/2026-09-02 2026-09-02T00:00:00Z Unbiased changelog entry for September 2, 2026. Product updates: September 1, 2026 tag:unbiased.ai,2026:changelog/2026-09-01 2026-09-01T00:00:00Z Unbiased changelog entry for September 1, 2026. Product updates: August 31, 2026 tag:unbiased.ai,2026:changelog/2026-08-31 2026-08-31T00:00:00Z Unbiased changelog entry for August 31, 2026. Verbosity Index Edition 2: The Tokenizer Confound https://unbiased.ai/blog/verbosity-index-edition-2/ 2026-08-25T00:00:00Z 72 fresh metered runs across six models, now measuring characters as well as tokens. Edition 1 had a confound, and correcting it changes the ranking.