Hunyuan Hy4 Preview vs MAI-Thinking-1

Tencent · China  |  Microsoft · US · Updated June 2026

Quick verdict

Pick Hunyuan Hy4 Preview for gpqa diamond (92.3) or terminal-bench (85.4). Pick MAI-Thinking-1 for very strong math reasoning (aime 2025 97%, aime 2026 94.5%) or microsoft's first in-house flagship reasoner, trained without openai distillation. Choose Hunyuan Hy4 Preview if you need self-hosting or data privacy; MAI-Thinking-1 if you want a managed API.

Hunyuan Hy4 Preview (Tencent, China) and MAI-Thinking-1 (Microsoft, US) line up two different AI ecosystems against each other — a comparison that is as much about cost philosophy and openness as raw capability. Hunyuan Hy4 Preview is tencent's open-sourced (Apache 2.0) next-generation Hunyuan flagship preview, released August 28, 2026 as a 770B-parameter (49B active) MoE model that helped optimize its own training pipeline. MAI-Thinking-1 is microsoft's first fully in-house flagship reasoning model — a Claude-class reasoner built independently to cut its OpenAI dependence. They diverge most on price, context window and open vs. closed weights — each quantified below from the models' real specs.

Key differences at a glance

Side-by-side specs

SpecHunyuan Hy4 PreviewMAI-Thinking-1
ProviderTencent (China) Microsoft (US)
ReleasedAugust 28, 2026 August 12, 2026
Context window1M+ tokens (~1,500 pages) 256K (~384 pages)
Price (in/out)$0.834/$2.501 per 1M tokens Not published
Open weight?Yes — self-hostable No — API only
Modalitiestext text, code
SWE-Bench VerifiedNot published Not published
MRCR v2 @ 1MNot published Not published

Who wins what

GPQA Diamond (92.3)

Hunyuan Hy4 Preview

Tencent's open-sourced (Apache 2.0) next-generation Hunyuan flagship preview, released August 28, 2026 as a 770B-parameter (49B active) MoE model that helped optimize its own training pipeline — and it carries the larger 1M+ tokens context.

Terminal-Bench (85.4)

Hunyuan Hy4 Preview

MAI-Thinking-1 is comparatively weak here — benchmarks are largely self-reported

SWE-bench Multilingual (82.9)

Hunyuan Hy4 Preview

Tencent's open-sourced (Apache 2.0) next-generation Hunyuan flagship preview, released August 28, 2026 as a 770B-parameter (49B active) MoE model that helped optimize its own training pipeline — and its weights are open while MAI-Thinking-1 is API-only.

Very strong math reasoning (AIME 2025 97%, AIME 2026 94.5%)

MAI-Thinking-1

MAI-Thinking-1 lists very strong math reasoning (AIME 2025 97%, AIME 2026 94.5%) among its strengths; Hunyuan Hy4 Preview does not.

Microsoft's first in-house flagship reasoner, trained without OpenAI distillation

MAI-Thinking-1

Hunyuan Hy4 Preview is comparatively weak here — labeled a 'preview', not yet Tencent's finalized GA flagship

Efficient reasoning at low token cost for its class

MAI-Thinking-1

MAI-Thinking-1 lists efficient reasoning at low token cost for its class among its strengths; Hunyuan Hy4 Preview does not.

Lowest cost at scale

MAI-Thinking-1

Its weights are open, so at volume you pay for your own hardware instead of Hunyuan Hy4 Preview's $0.834/$2.501 per 1M tokens.

Largest single-prompt input

Hunyuan Hy4 Preview

Its 1M+ tokens window is about 3.9× larger than MAI-Thinking-1's 256K, fitting roughly 1,500 pages in one prompt.

Which should you pick?

A cost-sensitive startup shipping high volume

MAI-Thinking-1

At Not published it undercuts Hunyuan Hy4 Preview, and on millions of tokens that margin decides the monthly bill.

Someone analysing very long documents or codebases

Hunyuan Hy4 Preview

Larger 1M+ tokens window fits more in one prompt.

A team with data-privacy or self-hosting needs

Hunyuan Hy4 Preview

Open weights let you run it on your own hardware; MAI-Thinking-1 is API-only.

Anyone whose priority is gpqa diamond (92.3)

Hunyuan Hy4 Preview

It is specifically built for that.

Anyone whose priority is very strong math reasoning (aime 2025 97%, aime 2026 94.5%)

MAI-Thinking-1

That is its strongest area.

An enterprise with regional data-residency rules

MAI-Thinking-1 or Hunyuan Hy4 Preview

Origin (China vs US) affects where data is processed and which compliance regime applies — check the provider's terms for your region.

Hunyuan Hy4 Preview: where it fits

Tencent's open-sourced (Apache 2.0) next-generation Hunyuan flagship preview, released August 28, 2026 as a 770B-parameter (49B active) MoE model that helped optimize its own training pipeline. Released August 28, 2026 by Tencent, it is built for gPQA Diamond (92.3), terminal-Bench (85.4), sWE-bench Multilingual (82.9), and large jump in long-horizon coding vs prior Hunyuan model (DeepSWE 28.0 to 64.3).

Its trade-offs are real: text-only, no native vision support, labeled a 'preview', not yet Tencent's finalized GA flagship, and sWE-Marathon score still low in absolute terms (31.9) despite a large relative jump. At $0.834 in / $2.501 out per million tokens, it sits in the budget price band.

MAI-Thinking-1: where it fits

Microsoft's first fully in-house flagship reasoning model — a Claude-class reasoner built independently to cut its OpenAI dependence. Released August 12, 2026 by Microsoft, it is built for very strong math reasoning (AIME 2025 97%, AIME 2026 94.5%), microsoft's first in-house flagship reasoner, trained without OpenAI distillation, efficient reasoning at low token cost for its class, and competitive with Claude Opus 4.6 on SWE-Bench Pro (vendor-reported).

Its trade-offs: closed and in private preview — no open weights, no published pricing, thin availability, and benchmarks are largely self-reported.

The bottom line for this matchup

The defining split here is open vs. closed. Hunyuan Hy4 Preview gives you weights you control — self-host it, fine-tune it, keep data in-house, pay only for hardware. MAI-Thinking-1 gives you a managed, always-updated API with no infrastructure to run. Teams with GPUs, privacy requirements, or huge volume often favour the open model; teams that want zero ops and the latest capabilities favour the closed one. Capability is close enough that this operational question, not the benchmark, usually decides it.

Want both Hunyuan Hy4 Preview and MAI-Thinking-1 without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.

See pricing

Frequently asked questions

Is Hunyuan Hy4 Preview or MAI-Thinking-1 better for coding?

Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Hunyuan Hy4 Preview leans toward gpqa diamond (92.3) while MAI-Thinking-1 leans toward very strong math reasoning (aime 2025 97%, aime 2026 94.5%), and that positioning usually predicts which feels better on your codebase.

Which is cheaper, Hunyuan Hy4 Preview or MAI-Thinking-1?

Hunyuan Hy4 Preview is open-weight, so self-hosting means no per-token fee (you pay for hardware instead), while MAI-Thinking-1 is API-metered at Not published. For most teams without GPUs, the API model is cheaper to start; at very high volume, self-hosting can win.

Which has the bigger context window?

Hunyuan Hy4 Preview — 1M+ tokens vs 256K, about 3.9× larger. Useful only if the model actually reasons over the full window, which not all do.

Can I use both Hunyuan Hy4 Preview and MAI-Thinking-1 together?

Yes — a multi-model platform like LumiChats gives you Hunyuan Hy4 Preview, MAI-Thinking-1 and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.

Which is newer, Hunyuan Hy4 Preview or MAI-Thinking-1?

Hunyuan Hy4 Preview — released August 28, 2026, about 16 days after MAI-Thinking-1.

Related comparisons

Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.