Revolut moves to self-hosted GLM
Summary
Revolut moved ~2,000 engineers onto self-hosted GLM models for coding, retiring Claude Code and Codex to an exception list — running on 56 Nvidia B200s at a blended 30 cents per million tokens (about a tenth of prior Anthropic/OpenAI spend), with engineering lead time improving 70%.
View Cached Full Text
Cached at: 11/03/26, 12:43 PM
BREAKING: Revolut moved ~2,000 engineers onto self-hosted GLM for coding and pushed Claude Code and Codex to the exception list.
Per David Abitbol, who leads AI ops there: 56 Nvidia B200s, a blended 30 cents per million tokens, roughly a tenth of what they paid Anthropic and OpenAI. Engineering lead time improved 70%.
Here’s the part that matters: the gains showed up when tokens stopped being rationed. Engineers quit saving AI for the hard tickets and pointed it at everything.
Frontier models become the exception, open weights become the default. That’s a pricing problem for every lab heading to IPO.
h/t @OliverMolander
Would you move your team off frontier APIs if the bill dropped 90%?
Similar Articles
@DeRonin_: My current local AI setup: - 2x DGX Spark linked (256gb) > GLM 5.2 @ 2bit, reasoning + agent loops - Mac Studio M3 Ultr…
A user describes their fully local AI stack using multiple hardware devices running Chinese models like GLM, Qwen, and Kimi, claiming 87% cost savings compared to frontier models like GPT-5.5 and Opus 4.8, while noting plans to self-host video generation.
@akshay_pachaar: Massive breakthrough here! Self-hosting LLMs just got ~75% cheaper: Most agent pipelines now run 4-5 small models under…
Superlinked releases SIE, an open-source inference engine that serves 85+ models behind one API with on-demand loading and LRU eviction, cutting self-hosting GPU costs by ~75% for agent pipelines.
@rohanpaul_ai: Surprising and such a good news for open source coding model, and also that there are lots of hidden chances to reduce …
Databricks tested GLM-5.2, an open-source coding model, and found it competes with top closed models like Claude Opus 4.8 on real enterprise code tasks while being cheaper ($1.28/task vs $1.94/task). The evaluation also highlighted Pi, a harness that reduces costs by sending less context per turn.
@0xSero: GLM-5.1-478B-NVFP4 Running on: - 4x RTX Pro 6000 - Sglang - 370,000 max tokens (1.75x full context) - p10 27.7 | p90 45…
A quantized 478B-parameter GLM-5.1 model runs on 4×RTX Pro 6000 GPUs via SGLang, delivering 370k-token context at up to 45 tok/s decode and 1340 tok/s prefill, and is demoed driving Figma.
@tolak_eth: I wanted to share how we avoided spending roughly $160k/year to host GLM-5.2 with its full 1M context. When GLM-5.2 lau…
Phala avoided $160k/year hosting costs for GLM-5.2 with full 1M context by quantizing MoE experts to 4-bit and keeping critical parts in FP8/BF16, achieving the same benchmark results on a single 8×H200 node and releasing the optimized model GLM-5.2-W4AFP8 on Hugging Face.