61–70 of 243 posts
Earlier quoted context omitted.
> that the comparisons would be extremely unfavorable. Why should they compare apples to oranges? Ministral3 Large costs ~1/10th of Sonnet 4.5. They clearly target different users. If you want a coding assistant you probably wouldn't choose this model for various reasons. There is place for more than only the benchmark king.
Come on. Do you just not read posts at all?
Earlier quoted context omitted.
Here's what I understood from the blog post: - Mistral Large 3 is comparable with the previous Deepseek release. - Ministral 3 LLMs are comparable with older open LLMs of similar sizes.
And implicit in this is that it compares very poorly to SOTA models. Do you disagree with that? Do you think these Models are beating SOTA and they did not include the benchmarks, because they forgot?
I am not sure why Meta paid 13B+ to hire some kid vs just hiring back or acquiring these folks. They'll easily catch up.
It's a good thing that open source models use the best arch available. K2 does the same but at least mentions "Kimi K2 was designed to further scale up Moonlight, which employs an architecture similar to DeepSeek-V3".
---
vllm/model_executor/models/mistral_large_3.py
```
from vllm.model_executor.models.deepseek_v2 import DeepseekV3ForCausalLM
class MistralLarge3ForCausalLM(DeepseekV3ForCausalLM):
```
"Science has always thrived on openness and shared discovery." btw
Okay I'll stop being snarky now and try the 14B model at home. Vision is good additional functionality on Large.
Earlier quoted context omitted.
That's unfair to Europe. A bunch of AI work is done in London (Deepmind is based here for a start)
Thats not the point. Deepmind is not an UK company, its google aka US. Mistral is a real EU based company.
Geometric mean of MMMLU + GPQA-Diamond + SimpleQA + LiveCodeBench : - Gemini 3.0 Pro : 84.8 - DeepSeek 3.2 : 83.6 - GPT-5.1 : 69.2 - Claude Opus 4.5 : 67.4 - Kimi-K2 (1.2T) : 42.0 - Mistral Large 3 (675B) : 41.9 - Deepseek-3.1 (670B) : 39.7 The 14B 8B & 3B models are SOTA though, and do not have chinese censorship like Qwen3.
How is there such a gap between Gemini 3 vs GPT 5.1/Opus 4.5? What is Gemini 3 crushing the others on?
Earlier quoted context omitted.
All thanks to the US VCs that acutally have money to fund Mistral's entire business. Had they gone to the EU, Mistral would have gotten a miniscule grant from the EU to train their AI models.
1. so what 2. asml
2. Did ASML invest in Mistral in their first round of venture funding or was it US VCs all along that took that early risk and backed them from the very start?
Risk aversion is in the DNA and in almost every plot of land in Europe such that US VCs saw something in Mistral before even the european giants like ASML did.
ASML would have passed on Mistral from the start and Mistral would have instead begged to the EU for a grant.
I use large language models in http://phrasing.app to format data I can retrieve in a consistent skimmable manner. I switched to mistral-3-medium-0525 a few months back after struggling to get gpt-5 to stop producing gibberish. It's been insanely fast, cheap, reliable, and follows formatting instructions to the letter. I was (and still am) super super impressed. Even if it does not hold up in benchmarks, it still out…
On the API side of things my experience is that the model behaving as expected is the greatest feature.
There I also switched to Openrouter instead of paying directly so I can use whatever model fits best.
The recent buzz about ad-based chatbot services is probably because the companies no longer have an edge despite what the benchmarks say, users are noticing it and cancel paid plans. Just today OpenAI offered me 1 month free trial as if I wasn’t using it two months ago. I guess they hope I forget to cancel.
Earlier quoted context omitted.
The lack of the comparison (which absolutely was done), tells you exactly what you need to know.
They're comparing against open weights models that are roughly a month away from the frontier. Likely there's an implicit open-weights political stance here. There are also plenty of reasons not to use proprietary US models for comparison: The major US models haven't been living up to their benchmarks; their releases rarely include training & architectural details; they're not terribly cost effective; they often fail…
We’re actually at a unique point right now where the gap is larger than it has been in some time. Consensus since the latest batch of releases is that we haven’t found the wall yet. 5.1 Max, Opus 4.5, and G3 are absolutely astounding models and unless you have unique requirements some way down the price/perf curve I would not even look at this release (which is fine!)