RivenGet started

Blog

News, engineering deep dives, and guides from the Riven team.

GuidesJuly 26, 2026
The cheapest LLM APIs in 2026: real per-token math
Compare real per-token costs across budget and frontier LLM APIs in 2026. We break down input and output pricing per million tokens with honest math.
GuidesJuly 26, 2026
LLM API pricing comparison: how per-token billing really works
Understand how per-token LLM API pricing works, spot hidden gateway markups, and see a worked monthly bill example comparing models at list price.
GuidesJuly 26, 2026
Self-hosted LLM platform: single-tenant AI for enterprises
Learn what self-hosted single-tenant LLM platforms offer, who needs them, and how hybrid deployments combine owned GPUs with cloud API failover.
ProductJuly 25, 2026
Kimi K3 is live on Riven
Kimi K3 — 2.8T parameters, 1M-token context, #1 on LMArena code arena in 24 hours — is live on Riven at Moonshot list price: $3 in, $15 out per million tokens, transparent per-token pricing, OpenAI-compatible.
GuidesJuly 25, 2026
GLM-5.2 in production: what it is good at
GLM-5.2 is a mixture-of-experts flagship strong at code, multilingual work, and cost-efficient production use. It is live on Riven with a full model guide.
GuidesJuly 25, 2026
Kimi K3 vs Claude Fable 5: which frontier model fits your workload
A practical comparison of Kimi K3 and Claude Fable 5 on context, coding, agentic use, and price per million tokens. Both are live on Riven at list price.
GuidesJuly 25, 2026
One API, 160+ models: the case for a unified gateway
An honest comparison of LLM aggregator pain points — rate limits, discounted routes, model masking — versus one governed gateway at vendor list price.
GuidesJuly 25, 2026
Pay-as-you-go LLM pricing, explained
Pay-as-you-go LLM pricing on Riven: $3 per 1M tokens on standard lanes, $0 to start, no seat minimums, and how per-token metering beats a subscription.
EnterpriseJuly 25, 2026
What single-tenant AI actually means
What single-tenant AI really means in practice: data isolation, compliance posture, self-hosted versus shared SaaS, and governance through Riven Control.
EngineeringJune 6, 2026
Agents with operational memory
How Riven's internal agents keep durable operational memory with MemPalace to search platform state and remember context across sessions for reliability.
EngineeringJune 3, 2026
Pricing Brain v1: how Riven prices 160+ models honestly
Inside Pricing Brain v1, the internal system that tracks 65 model vendor list rates and competitor prices daily so Riven's datasheet stays current and honest.
ProductMay 22, 2026
Riven Chat v2 is live
Riven Chat v2 is live on Riven Auth with OIDC, replacing the legacy VM stack. Registration, callback, session, and redirect flows verified end to end.
EnterpriseMay 15, 2026
The self-hosted deployment guide
How enterprises run Riven single-tenant on their own infrastructure: containerized services, reverse proxy edge, private networking, governed by Riven Control.