FinOps for LLM – Leanroute Team wrote a blog

1•leanroute_ai•52m ago

I found this is probably the simplest and realistic (no fluff) take on FinOps for LLMs. https://leanroute.dev/blog/finops-for-llm-spend-three-axes-of-cost-variance

Comments

MoneyLovesSpeed•48m ago

the part about lower spend not always meaning better optimization is pretty important, sometimes you just had less traffic lol

leanroute_ai•39m ago

exactly, lost the user-base but kept thanking the code being more efficient lol

The Telescope NASA Launched Sunday Was Built to Look at You

The Gemini app is now available for Windows

Thinking with Looped Flows

Metamorphosis: Transformations in the Early Modern Flap Book (19th Century)

One company that steals accuses another company that steals [pdf]

Show HN: Hekla, an event sourcing runtime with its own language

DeepSeek Harness

It Breaks a Village: Bevy's 6th Birthday

GPT Image 2.5 for Consulting Presentations: Faster Slides and Smarter Control

The Bayeux Tapestry: Woven by the Victors

How to Make a PwC-Style PowerPoint with AI (2): 5 More Consulting Frameworks

Trial Testing Out-of-Body Experiences Produces Unexpected Twist

Dioxus Labs is joining Cognition

9/11 in Realtime (synced globally for 9/11)

Günther Anders, the Philosopher at the End of the World

Outflow connects directly to inverters, mints solar electricity records

Show HN: CertArena – Free AI certification simulator (zero deps, native Node 24)

Reconstructing Concurrency Invariants Through Medieval East Asian Logic

LiteLLM Alternatives?

AI Is Killing the App Economy and Building Its Successor

AI Thumbnail Generator for YouTube – Try Free – Thumbrix

Solving Alignment with a Prompt

Whose actions are these proposals intended to constrain?

Apple A20 Pro Geekbench Result

California enacts new curbs on social media for children

FinOps for LLM – Leanroute Team wrote a blog

Serve Qwen3 2.4T-Parameter Model with Configurable Reasoning on Nvidia GB300

OpenAI's Astra Made It to a Nether Fortress in Minecraft

Front-End Shouldn't Know What Marketing Wants

Ask HN: Those running agents 24/7, what's your workflow and what are they doing?

FinOps for LLM – Leanroute Team wrote a blog

Comments

The Telescope NASA Launched Sunday Was Built to Look at You

The Gemini app is now available for Windows

Thinking with Looped Flows

Metamorphosis: Transformations in the Early Modern Flap Book (19th Century)

One company that steals accuses another company that steals [pdf]

Show HN: Hekla, an event sourcing runtime with its own language

DeepSeek Harness

It Breaks a Village: Bevy's 6th Birthday

GPT Image 2.5 for Consulting Presentations: Faster Slides and Smarter Control

The Bayeux Tapestry: Woven by the Victors

How to Make a PwC-Style PowerPoint with AI (2): 5 More Consulting Frameworks

Trial Testing Out-of-Body Experiences Produces Unexpected Twist

Dioxus Labs is joining Cognition

9/11 in Realtime (synced globally for 9/11)

Günther Anders, the Philosopher at the End of the World

Outflow connects directly to inverters, mints solar electricity records

Show HN: CertArena – Free AI certification simulator (zero deps, native Node 24)

Reconstructing Concurrency Invariants Through Medieval East Asian Logic

LiteLLM Alternatives?

AI Is Killing the App Economy and Building Its Successor

AI Thumbnail Generator for YouTube – Try Free – Thumbrix

Solving Alignment with a Prompt

Whose actions are these proposals intended to constrain?

Apple A20 Pro Geekbench Result

California enacts new curbs on social media for children

FinOps for LLM – Leanroute Team wrote a blog

Serve Qwen3 2.4T-Parameter Model with Configurable Reasoning on Nvidia GB300

OpenAI's Astra Made It to a Nether Fortress in Minecraft

Front-End Shouldn't Know What Marketing Wants

Ask HN: Those running agents 24/7, what's your workflow and what are they doing?