Model comparison
Hy3 Preview vs MiniMax M3
Head-to-head evidence from 16 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Verified leaderboard positions: Hy3 Preview unranked; MiniMax M3 #18
BenchAlign evidence: Hy3 Preview not scored; MiniMax M3 supported. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Hy3 Preview and MiniMax M3 share 16 comparable benchmark results. 2 of 8 categories are comparable. 6 results are unique to Hy3 Preview; 29 to MiniMax M3.
Updated July 16, 2026- Shared results
- 16
- Hy3 Preview only
- 6
- MiniMax M3 only
- 29
- Comparable categories
- 2 / 8
Pick MiniMax M3 if you want the stronger benchmark profile. Hy3 Preview only becomes the better choice if you want the cheaper token bill or you want the stronger reasoning-first profile.
Confidence note. This is a partial-evidence comparison with 16 shared benchmark results across 4 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
MiniMax M3 is clearly ahead on the provisional aggregate, 70 to 53. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
MiniMax M3's sharpest advantage is in agentic, where it averages 72.3 against 54.4. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 54.4% to 66%.
MiniMax M3 is also the more expensive model on tokens at $0.30 input / $1.20 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Hy3 Preview. That is roughly Infinityx on output cost alone. Hy3 Preview is the reasoning model in the pair, while MiniMax M3 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. MiniMax M3 gives you the larger context window at 1M, compared with 256K for Hy3 Preview.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Hy3 Preview | Δ | MiniMax M3 |
|---|---|---|---|
| Agentic | Hy3 Preview54.4 | Margin→ 17.9 | MiniMax M372.3 |
| Coding | Hy3 Preview57.8 | Margin→ 14.4 | MiniMax M372.2 |
| Knowledge | Hy3 Preview33.8 | MarginNo overlap | MiniMax M3Not measured |
| Math | Hy3 PreviewNot measured | MarginNo overlap | MiniMax M385.7 |
| Multimodal | Hy3 PreviewNot measured | MarginNo overlap | MiniMax M364.9 |
| Inst. Following | Hy3 Preview63.1 | MarginNo overlap | MiniMax M3Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 54.4%B 66%Winner: MiniMax M3Δ 11.6Terminal-Bench 2.0: Hy3 Preview scored 54.4%; MiniMax M3 scored 66%. MiniMax M3 wins this benchmark. - Source ↗
SWE-bench Verified
CodingA 74.4%B 80.5%Winner: MiniMax M3Δ 6.1SWE-bench Verified: Hy3 Preview scored 74.4%; MiniMax M3 scored 80.5%. MiniMax M3 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Hy3 Preview | MiniMax M3 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Hy3 Preview$0 input / $0 output | MiniMax M3$0.3 input / $1.2 output | Hy3 Preview has the lower combined listed price. |
| Generation speedtokens per second | Hy3 PreviewNot available | MiniMax M3Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Hy3 PreviewNot available | MiniMax M3Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Hy3 Preview256K | MiniMax M31M | MiniMax M3 lists the larger context window. |
Benchmark Deep Dive
AgenticMiniMax M3 wins17 benchmarks
| Benchmark | Hy3 Preview | MiniMax M3 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 54.4% | 66% | MiniMax M3 leads |
| Gert LabsSource | 36.91% | — | Not comparable |
| AA Agentic IndexSource | 30.7% | 35.4% | MiniMax M3 leads |
| GDPval-AASource | 35.7% | 44.7% | MiniMax M3 leads |
| GDPval-AASource | 1214 | 1395 | MiniMax M3 leads |
| BrowseCompSource | — | 83.5% | Not comparable |
| OSWorld-VerifiedSource | — | 70.1% | Not comparable |
| MCP AtlasSource | — | 74.2% | Not comparable |
| Claw-EvalSource | — | 74.5% | Not comparable |
| τ²-bench resultsSource | — | 88.9% | Not comparable |
| GDPval rubricsSource | — | 74.7% | Not comparable |
| BankerToolBenchSource | — | 76.1% | Not comparable |
| ResearchClawBenchSource | — | 19.8% | Not comparable |
| OSWorld 2.0Source | — | 4.6% | Not comparable |
| AA BriefcaseSource | — | 1110 | Not comparable |
| AA EnterpriseOps-GymSource | — | 32.1% | Not comparable |
| AA Harvey LABSource | — | 6.7% | Not comparable |
CodingMiniMax M3 wins12 benchmarks
| Benchmark | Hy3 Preview | MiniMax M3 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 74.4% | 80.5% | MiniMax M3 leads |
| Terminal-Bench 2.0Source | 54.4% | 66.0% | MiniMax M3 leads |
| SciCodeSource | 41.2% | — | Not comparable |
| AA-SciCodeSource | 47.6% | 45.4% | Hy3 Preview leads |
| AA Coding IndexSource | 58.8% | 58.6% | Hy3 Preview leads |
| SWE-bench ProSource | — | 59% | Not comparable |
| NL2RepoSource | — | 42.1% | Not comparable |
| Terminal-Bench HardSource | — | 42.4% | Not comparable |
| VIBE V2Source | — | 50.1% | Not comparable |
| SVG-BenchSource | — | 63.7% | Not comparable |
| KernelBench HardSource | — | 28.8% | Not comparable |
| AA Terminal-Bench 2.1Source | — | 65.2% | Not comparable |
Reasoning2 benchmarks
Knowledge10 benchmarks
| Benchmark | Hy3 Preview | MiniMax M3 | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 41.2% | 44.4% | MiniMax M3 leads |
| GPQASource | 87.2% | — | Not comparable |
| GPQA-DSource | 87.2% | — | Not comparable |
| HLESource | 25.5% | — | Not comparable |
| AA-Omniscience AccuracySource | 31.5% | 15.0% | Hy3 Preview leads |
| AA-Omniscience Hallucination RateSource | 73.0% | 16.1% | MiniMax M3 leads |
| AA-GPQA DiamondSource | 89.7% | 92.9% | MiniMax M3 leads |
| AA-HLESource | 31.6% | 37.1% | MiniMax M3 leads |
| AA-Omniscience IndexSource | -18.5% | 1.4% | MiniMax M3 leads |
| AA Openness IndexSource | — | 33.3% | Not comparable |
Math1 benchmarks
| Benchmark | Hy3 Preview | MiniMax M3 | Result |
|---|---|---|---|
| USAMO 2026Source | — | 85.7% | Not comparable |
Multimodal7 benchmarks
| Benchmark | Hy3 Preview | MiniMax M3 | Result |
|---|---|---|---|
| OfficeQA ProSource | — | 45.1% | Not comparable |
| OmniDocBench 1.5Source | — | 91.6% | Not comparable |
| MMMU-ProSource | — | 78.1% | Not comparable |
| VideoMMMUSource | — | 84.6% | Not comparable |
| Video-MME (with subtitle)Source | — | 85.4% | Not comparable |
| Design Arena WebsiteSource | — | 1294 | Not comparable |
| AA-MMMU-ProSource | — | 78.6% | Not comparable |
Frequently Asked Questions (3)
Which is better, Hy3 Preview or MiniMax M3?
MiniMax M3 is ahead on BenchLM's provisional leaderboard, 70 to 53. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 54.4% and 66%.
Which is better for coding, Hy3 Preview or MiniMax M3?
MiniMax M3 has the edge for coding in this comparison, averaging 72.2 versus 57.8. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Hy3 Preview or MiniMax M3?
MiniMax M3 has the edge for agentic tasks in this comparison, averaging 72.3 versus 54.4. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
The AI models change fast. We track them for you.
A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.