--- title: "LiteLLM's '7x faster' claim is missing from its own benchmark data; Stripe pays $8 billion for OpenRouter" description: "A line-by-line check of LiteLLM's own benchmark files finds no 7x gap, while Google sells routing decisions at $0.10 per million tokens and Stripe pays $8 billion for a router. The moat is governance, and it is bleeding." publisher: "The Inference" section: "Platforms" published: 2026-11-12T15:36:30.292Z modified: 2026-11-12T15:36:30.292Z canonical: https://theinference.org/article/litellm-s-7x-faster-claim-is-missing-from-its-own-benchmark-data-stripe-pays-8-billion-for-openrouter language: en keywords: "LiteLLM, OpenRouter, AI Gateways, Benchmarks, Stripe" --- # LiteLLM's '7x faster' claim is missing from its own benchmark data; Stripe pays $8 billion for OpenRouter > A line-by-line check of LiteLLM's own benchmark files finds no 7x gap, while Google sells routing decisions at $0.10 per million tokens and Stripe pays $8 billion for a router. The moat is governance, and it is bleeding. ## The headline, checked against the file it came from [LiteLLM](https://theinference.org/markets/companies/litellm), the open-source AI gateway behind an estimated 92 million PyPI downloads a month [4], sells its Rust rewrite on speed: 0.66 ms of added p99 latency, roughly 22 MB of memory, "4.5x more requests per dollar" than Bifrost [4]. [A third-party check published on November 11](https://dev.to/therealmrmumba/is-litellm-rust-really-7x-faster-a-closer-look-at-the-numbers-497m) went back to the raw CSV in LiteLLM's own ai-gateway-bench repository and found the "7x" headline traces to no row of the latency sweep [1]. The 4.5 ms Bifrost figure used in the marketing simply does not appear in the data [1]. ## Bifrost wins at 1 concurrent request and at 64 [The committed CSV](https://raw.githubusercontent.com/BerriAI/ai-gateway-bench/main/results/latency_vs_concurrency.csv), checked directly this week, confirms it [3]. At one concurrent request Bifrost adds 0.75 ms and Rust 0.89 ms: Bifrost wins [3]. At 64 concurrent requests Bifrost adds 28.68 ms and Rust 40.28 ms: Bifrost wins again [3]. Rust leads only at the middle points, by as little as 0.02 ms [3]. At the top tested load the two gateways sit 0.1% apart on throughput, 2,722 versus 2,719 requests a second [3]. The dollar figure is a model built on assumed cloud prices, not a measured bill [2]. The Rust port itself is 3.11% of the repository by source bytes, with 24% of major APIs built [1]. *Chart: **No 7x gap: Bifrost beats LiteLLM Rust at both ends of the sweep** Line chart of added p99 latency in milliseconds for LiteLLM Rust and Bifrost at 1, 4, 16 and 64 concurrent requests. Bifrost is lower at 1 and at 64; Rust is lower at 4 and 16 by 0.02 to 0.56 milliseconds. A reference line marks the 0.66 millisecond figure LiteLLM markets.* *Added p99 latency in milliseconds at each concurrency level in BerriAI's committed results/latency_vs_concurrency.csv, lower is better. The gateways cross twice: Rust leads by 0.02 ms at 4 and 0.56 ms at 16, and trails at 1 and at 64. The reference line marks the 0.66 ms added p99 that LiteLLM's homepage sells. Sources: ai-gateway-bench CSV, LiteLLM homepage. [3][4]* ## Google prices the routing layer from above The same week, Google began selling routing decisions directly: [Gemini Decision 1 scores up to 1,000 closed-set questions per pass](https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/) at $0.10 per million input tokens, zero output tokens billed, and names LiteLLM among its rolling integrations [5]. [Stripe](https://theinference.org/markets/companies/stripe), meanwhile, [agreed to pay $8 billion for OpenRouter](https://www.businessinsider.com/nvidia-spacex-deals-fuel-historic-boom-ai-startup-manda-2026-10), a routing business whose own measured overhead in LiteLLM's benchmark was 61 ms at p50 [2][6]. Eight billion dollars for 61 ms says buyers are pricing governance and margin, not milliseconds [2][6]. ## The moat is governance, and it is bleeding What is hard to copy in LiteLLM is not sub-millisecond forwarding; it is the price list, budgets and audit trail. That is also its attack surface: [the March PyPI compromise](https://blog.gitguardian.com/inside-the-litellm-hack/) exfiltrated 153 GB across 433,909 files and 2,488 organizations [7], and October's malicious 1.107.0 release came from a deleted-but-unrevoked admin key that stayed valid for four months [7]. Watch Bifrost's promised Part 2 rerun [1], and whether the November full-feature lane, where Rust does lead [2], survives outside replication. ## Takeaway LiteLLM's "7x faster" headline traces to no row in its own committed benchmark, and at $8 billion for OpenRouter buyers are pricing routing governance, not milliseconds. ## Sources 1. [dev.to (therealmrmumba), Is LiteLLM Rust Really 7x Faster? A Closer Look at the Numbers, 12 November 2026](https://dev.to/therealmrmumba/is-litellm-rust-really-7x-faster-a-closer-look-at-the-numbers-497m) 2. [BerriAI, ai-gateway-bench repository README (full-feature results, cost model, hosted OpenRouter arm), updated 11 November 2026](https://github.com/BerriAI/ai-gateway-bench) 3. [BerriAI, ai-gateway-bench results/latency_vs_concurrency.csv](https://raw.githubusercontent.com/BerriAI/ai-gateway-bench/main/results/latency_vs_concurrency.csv) 4. [LiteLLM homepage, benchmark claims section, checked 12 November 2026](https://www.litellm.ai/) 5. [Google DeepMind blog, Gemini Decisions API: Fast, calibrated multimodal decisions, 12 November 2026](https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/) 6. [Business Insider, VC Insider: Nvidia, SpaceX deals fuel historic boom in AI startup M&A, October 2026](https://www.businessinsider.com/nvidia-spacex-deals-fuel-historic-boom-ai-startup-manda-2026-10) 7. [GitGuardian, Inside the LiteLLM hack: 153GB, 433,909 Files, 2,488 Organizations, 6 November 2026](https://blog.gitguardian.com/inside-the-litellm-hack/) --- LiteLLM's '7x faster' claim is missing from its own benchmark data; Stripe pays $8 billion for OpenRouter — The Inference. Canonical: https://theinference.org/article/litellm-s-7x-faster-claim-is-missing-from-its-own-benchmark-data-stripe-pays-8-billion-for-openrouter