ResearchLLM Verifications Leaderboard

Data as of · 78 models, 80 questions · reviewed by Suhas Bhairav · methodology

Head to head · tested 27 Oct 2026

Hy3 vs MiniMax M3: accuracy, speed and price

Similar accuracy; Hy3 costs less per question.

TencentHy358.8% score · #41$0.0012 per question96 tok/sMiniMaxMiniMax M357.5% score · #43$0.0026 per question121 tok/s

Verdict

Built from the measurements below

  • Accuracy is too close to call: Hy3 scored 58.8% and MiniMax M3 57.5% on the same 80 questions, a gap within what chance alone could produce.
  • Hy3 is 2.1× cheaper per question as billed ($0.0012 vs $0.0026), reasoning tokens included.
  • MiniMax M3 generates text 1.3× faster: 121 tokens per second against 96.
  • MiniMax M3 starts answering sooner: 0.5 s to the first token, against 1.7 s for Hy3.
  • MiniMax M3 accepts more context: 1M tokens against 262k.

Which should you choose?

Choose Hy3 for the lower cost per correct answer ($0.0021 against $0.0045). Choose MiniMax M3 for faster streaming, a quicker first token, a 1M-token context window and more hosts to choose from on OpenRouter (12 against 5).

Accuracy gap +1.3 points (z = 0.16). A winner is called only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.

Head to head

Same questions, same prompt, same API

MeasureHy3MiniMax M3Edge
Verification scoreHigher is better58.8%57.5%Too close to call
Cost per questionLower is better$0.0012$0.0026Hy3
Output speedHigher is better96121MiniMax M3
Time to first tokenLower is better1.7 s0.5 sMiniMax M3
End-to-end response timeLower is better9.0 s4.2 sMiniMax M3
Tokens per questionLower is better2,3552,129MiniMax M3
Context windowHigher is better262k1MMiniMax M3

Side by side

Bars are scaled to the larger of the two values. The winner on each measure is marked in bold.

Every measure

Hy3 first, MiniMax M3 second

Verification scorehigher is better
Hy358.8%
MiniMax M357.5%
Cost per questionlower is better
Hy3$0.0012
MiniMax M3$0.0026
Output speedhigher is better
Hy396
MiniMax M3121
Time to first tokenlower is better
Hy31.7 s
MiniMax M30.5 s
End-to-end response timelower is better
Hy39.0 s
MiniMax M34.2 s
Tokens per questionlower is better
Hy32,355
MiniMax M32,129
Context windowhigher is better
Hy3262k
MiniMax M31M

Score vs. cost per question

Both models in color, every other model in grey

Most attractive quadrant0255075100$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Verification score (%) →Qwen3.8 2.4T A95BCommand AReka Flash 3Gemma 4 26B A4BGemma 4 31BQwen3.5 397B A17BHermes 4 405BMuse WonderGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Phi 4Granite 4.2 8BHy3GLM 5.3 FlashSolar Pro 4Trinity Large ThinkingMiniMax M3Mistral Large 4Claude Opus 5.5GPT-6.1 SolGemini 4 ArgonGPT-5.4GPT-6 Lunagpt-oss-120bLlama 4 Maverick

Specifications and pricing

From OpenRouter's catalogue, plus the leaderboard's test dates

SpecificationHy3MiniMax M3
CreatorTencentMiniMax
Released6 Jul 202631 May 2026
WeightsOpen weightsOpen weights
ReasoningYesYes
Inputtexttext, image, video
Outputtexttext
Context window262k tokens1M tokens
Knowledge cutoff——
List price in / out per 1M$0.132 / $0.528$0.3 / $1.2
Blended price per 1M (3:1)$0.23$0.52
Cheapest hostTencent · $0.0825 / $0.33DeepInfra · $0.224 / $0.88
Hosts on OpenRouter512
Last tested26 Oct 202626 Oct 2026

Compare any two models →

Hy3 vs MiniMax M3: frequently asked questions

Is Hy3 better than MiniMax M3?
On accuracy they are statistically tied: Hy3 scored 58.8% and MiniMax M3 57.5% on the same 80 questions from open datasets. Similar accuracy; Hy3 costs less per question.
Which is cheaper, Hy3 or MiniMax M3?
Hy3 is cheaper: $0.0012 per question as billed, against $0.0026. List prices: Hy3 $0.132 in / $0.528 out, MiniMax M3 $0.3 in / $1.2 out per million tokens.
Which is faster, Hy3 or MiniMax M3?
MiniMax M3 is faster at 121 output tokens per second, against 96. Time to first token: 1.7 s for Hy3, 0.5 s for MiniMax M3.
Which has the larger context window, Hy3 or MiniMax M3?
MiniMax M3, with 1M tokens against 262k.