Inception Mercury 2

“gets roflstomped by Mercury 3 at both cost and performance at similar TPS speed”

“It's not really all that fast, for most things it's full seconds or at least a significant portion of a second.”

“it's still incredibly fast compared to what I'm used to with other models”

“love these guys. instead of spending time on something akin to those creepy "dots"... they ship actual improvements to speed”

“It's really fast and works really well in my initial tests.”

“Those are some far better .svgs than I got last Mercury 2.5. Unfortunately no real improvement under iteration”

“early testing has found the model to be great for coding subagents”

“Customers are already using Mercury 3 for browser use agents, interactive education, real-time search agents, customer support, and more.”

“I'm going to start using this instead of Haiku/GPT-Luna for tasks that require speed.”

“I created a pelican worked wonderfully, better than old haiku model or even sonnet models”

“Having a cheap 1,000 tok/s option for coding work is a dream”

“Seems like a definite improvement over 2.5 though in my testing”

“Outperforms GPT-6 Luna and Claude Haiku 5.5 across a range of benchmarks at matched reasoning effort.”

“for small, transactional tasks it is quite strong and fast”

“Mercury 2.5 is below average in intelligence, but well priced when comparing to other models of similar price.”

“the lack of an input cache discount makes it prohibitively expensive for most use cases that aren't one-shot prompts”

“Pricing at $0.25 and $0.75 already puts its cost well above reasonably reputable inference providers”

“The speed means absolutely nothing when it is finishing almost dead last when compared to the frontier AI companies.”

“Mercury 2.5 is truly groundbreaking, giving it a try is highly recommended.”

“it is light years behind frontier models. Speed is great, but wrong is much worse than slow”

“with thinking off, it was considerably better than Mercury 2 and generally good in fact, not very sloppy”

“Even pretty dumb models are useful for running subtasks (especially at this sort of speed).”

“this model and previous are halfway decent and you can do diffusion sub-agents”

“I like the text output on logical and historical content that I sampled so far”

“having a model run at 1100tps goes a long a way to alleviate that”

“mercury-2 was pretty good”

“It's much less impressive-looking than any of the above, but it cost less than 1/20th of a cent”

“Mercury-2 is amazing. I am using it frequently as the arbiter in llm-consortium”

“The virtually instantaneous responses of Inception's Mercury models”

“Output quality, tool calling reliability and task adherence tend to be far more reliable across all three over Mercury 2”

“it made some interactions nearly frictionless, very close to how old school autocomplete/T9/autocorrect works”

“So Mercury 2 scores significantly higher on speed (96 vs 62), while being comparable on cost efficiency”

“it generates incorrect information, and it doesn't fix it when later in the text it realizes that it is incorrect”

“I had a very odd interaction somewhat similar to how weak transformer models get into a loop”

“We see Mercury 2 as enabling very fast iteration for agentic tasks.”

“The main differentiator is latency — it’s ~5× faster at comparable quality.”

“I can see some promise with diffusion LLMs, but getting them comparable to the frontier is going to require a ton of work”

“It seems like the chat demo is really suffering from the effect of everything going into a queue.”

“What excites me most about these new 4figure/second token models is that you can essentially do multi-shot prompting”

“The speed is still impressive when playing aroun with it sometimes”