Hunyuan Hy4 vs Mistral Large 4

Compare models

Hunyuan Hy4 is more worth it — 49% vs 47%

How we score worth-it? →
49%Worth it¥6/¥18 per 1M ($0.83/$2.50)Per-use

Tencent's open-source MoE flagship — 770B total/49B active params, 1M context, top open-source tier for math and agentic work.

✅ Pros (real reviews)
  • Hy4 apparently has ludicrous traction on OpenRouter already, with trillions of tokens processed in a couple days. Relatively cheap with a 5% cache cost.— hn
  • I'm very glad that models which are continuously improving in capability and getting closer to closed-source ones can be open-sourced.— huggingface
  • WTF! That's insane. Someone test this lol.— reddit ↗
  • The future of local models is brighter than expected. We can still compress them further — Hy4 has not yet been sufficiently post-trained.— reddit ↗
  • And this is a PREVIEW? The actual release must be fable level or higher then?— reddit ↗
⚠️ Cons (real reviews)
  • Maybe it's a problem with the hosting, but I didn't get much useful out of this model as a coding agent.— hn
  • I used CodeBuddy + Hy4 preview to explore a few things and it kept getting them wrong; switching to another model was the right call. I'll never trust Hy again.— v2ex ↗
  • The only saving grace from an onslaught of 'it overthinks' posts is that it's a huge chonker of a model, too big for most people.— reddit ↗
  • Aggressive quants don't degrade uniformly — short single-turn questions barely move, but long agentic runs fall apart around turn ten.— reddit ↗
  • Thinking takes too long with Hy4 — it thought for over 10 minutes just to change a bit of frontend code.— blog ↗

“Horizon Math score is the best among all open models, second only to GPT-5.6 Sol.”

“This is a PREVIEW? The actual release must be fable level or higher.”

“Thinking takes too long — it thought over 10 minutes just to change a bit of frontend code.”

47%Worth it$0.70/$2.10 per 1MPer-use

European open-weights flagship with aggressive preview pricing.

✅ Pros (real reviews)
  • What's more impressive is the datacenter used for training. It shows you can get "sub-SOTA" for one tenth of the resources of SOTA.— hn
  • It's quite fast— hn
  • "Covered by GDPR" is a very big plus for me. Also very simple training opt-out.— hn
  • High one is actually much better — the feet connect to the pedals, no hub cap, relatively proper position.— hn
  • It's 10x cheaper than Mistral Medium 3.5 from April and goes from 58% to 74% correct. Definitely a generational shift.— hn
⚠️ Cons (real reviews)
  • Everything feels very European. Unproductive company that is 3rd place... The only thing that gives it value is that it's generated in some region in Europe, like random cheese or wine. But I still can't imagine a usecase outside patriotism.— hn
  • I already ran it on my adversarial programming language benchmark a few weeks ago and it performed pretty poorly (on par with Kimi K3).— hn
  • The weird thing about this model is that it spits out Chinese characters just like the chinese models when chatting with it in German. Happened to me quite often during the past few days.— hn
  • The two pelicans are so similar it suggests the bird was deliberately fitted — we'd need other animals not part of a benchmark.— hn
  • Far from AGI. This shows a strong ability to generate a known benchmark image, not a measure of thought.— hn

“High-effort mode renders the pelican correctly — feet connect to pedals”

“Far from AGI — it just regenerated a well-known benchmark image”

“Aggressive preview pricing at $0.68/$2.09”

FAQ

Is Hunyuan Hy4 or Mistral Large 4 more worth it?

Hunyuan Hy4 is more worth it, with a 49% worth-it score vs 47% for Mistral Large 4.

How much does Hunyuan Hy4 cost vs Mistral Large 4?

Hunyuan Hy4: ¥6/¥18 per 1M ($0.83/$2.50) · Mistral Large 4: $0.70/$2.10 per 1M.

What do users say about Hunyuan Hy4 vs Mistral Large 4?

Hunyuan Hy4: Horizon Math score is the best among all open models, second only to GPT-5.6 Sol.. Mistral Large 4: High-effort mode renders the pelican correctly — feet connect to pedals.

Compare models →