OpenRouter (@OpenRouter) / X

OpenRouter

5,059 posts

OpenRouter

@OpenRouter

The largest LLM marketplace. 500+ models (incl. 50+ free), the best prices, explorable data, private chat, & a unified API. openrouter.ai/discord

Joined July 2023

Affiliates Replies Media

Pinned
OpenRouter
@OpenRouter
Jun 13
Introducing the Fusion API, the smartest compound model in the market. Fusion achieves Fable-level intelligence at half the price. How it works 👇
OpenRouter
@OpenRouter
Jul 9
GPT-5.6 is live on OpenRouter! Sol (flagship), Terra (balanced), and Luna (fast + low-cost). In our early testing, the 5.6 models were more token efficient, leading to lower cost and faster time to task completion. AA's results agree. More on Sol, Terra, and Luna below 🧵
OpenRouter
@OpenRouter
Jul 9
Replying to @OpenRouter
GPT-5.6 Terra: the balanced everyday driver. Coding, reasoning, and agentic tasks at GPT-5.5-competitive quality for roughly 2x lower cost.
GPT-5.6 Terra - API Pricing & Providers
From openrouter.ai
OpenRouter
@OpenRouter
Jul 9
GPT-5.6 Luna: the fast, low-cost tier. Tuned for high-volume, latency-sensitive work like chat, classification, and lightweight agents, with real GPT-5.6 reasoning.
GPT-5.6 Luna - API Pricing & Providers
From openrouter.ai
OpenRouter
@OpenRouter
Jul 8
Grok 4.5 is live now on OpenRouter! Cursor’s contributions have helped make @SpaceXAI’s model frontier-level on long-horizon engineering tasks.
SpaceXAI
@SpaceXAI
Jul 8
Announcing Grok 4.5, our first model trained specifically for coding and agents. It was trained with Cursor and offers frontier intelligence at leading speeds and cost efficiency. x.ai/news/grok-4-5
00:00
OpenRouter
@OpenRouter
Jul 8
Replying to @OpenRouter
Use it now:
Grok 4.5 - API Pricing & Providers
From openrouter.ai
OpenRouter
@OpenRouter
Jul 8
Read more:
Introducing Grok 4.5
From x.ai
OpenRouter
@OpenRouter
Jul 8
Use Grok 4.5 from @SpaceXAI the moment it goes live, with no code change. ~x-ai/grok-latest always routes to the newest Grok. Point at it now (Grok 4.3 today) and it auto-upgrades to 4.5 the second it ships. Try it:
Grok Latest - API Pricing & Benchmarks
From openrouter.ai
OpenRouter
@OpenRouter
Jul 8
New: one-click ZDR (zero-data retention) in the Chatroom Compare models side-by-side with full privacy: openrouter.ai/chat
OpenRouter reposted
Alex Atallah
@alexatallah
Jul 7
There are countless useful insights in OpenRouter data, which we make public to the world to improve everyone's use of intelligent models. DM me if you're obsessed with data and LLMs and want to be a part of something exciting coming up!
OpenRouter
@OpenRouter
Jul 7
We benchmarked 1,730 visual reasoning questions across 5 models to test a common cost-saving trick: setting image detail to "low." Surprise finding: it often backfires. The model burns extra reasoning tokens squinting at a blurry image, and overall cost increases 👇
OpenRouter
@OpenRouter
Jul 7
We benchmarked 1,730 visual reasoning questions across 5 models to test a common cost-saving trick: setting image detail to "low." Surprise finding: it often backfires. The model burns extra reasoning tokens squinting at a blurry image, and overall cost increases 👇
OpenRouter
@OpenRouter
Jul 7
Replying to @OpenRouter
So what actually moves the bill? Reasoning effort, by a wide margin. Detail swung accuracy 2 to 17 points while barely touching cost, while capping reasoning effort cut cost 50 to 75 percent with accuracy moving 1 to 2 points, within noise.
OpenRouter
@OpenRouter
Jul 7
TLDR: Reasoning models → keep detail on auto or high, control cost with reasoning effort Non-reasoning models → low detail genuinely saves money and latency if you can absorb the accuracy hit Full benchmark with per-model tables and methodology:
Choosing the Optimal Image Input Detail Level in LLMs — OpenRouter Blog
From openrouter.ai
OpenRouter reposted
Dasha
@dasha_shunina
Jul 6
every model choice is a bet. it shouldn't be a blind one. 750T+ tokens routed across every major model and provider — what actually moves cost, latency, and reliability. live at the top of salesforce tower w/ @deedydas (menlo) + @shashankgoyal95 (openrouter) rsvp 👇
OpenRouter
@OpenRouter
Jul 6
DeepSeek doubled its token share on OpenRouter in six months, going from 9% in January to 18% by June. The driver is agentic work. New data deep dive in our blog 👇
OpenRouter
@OpenRouter
Jul 6
Replying to @OpenRouter
V4 also broke DeepSeek away from the other Chinese models. At 6.6T weekly tokens it now runs well ahead of Xiaomi (4.0T), Minimax (3.6T), and Tencent (3.8T), and it edges out Anthropic's 6.1T as the highest-token-volume model lab on the platform.
OpenRouter
@OpenRouter
Jul 6
Full breakdown across 457 trillion tokens of request data here:
DeepSeek V4 Is Earning Agentic Token Share — OpenRouter Blog
From openrouter.ai