Model comparison
MiniMax M3 vs Step 3.7 Flash
Head-to-head evidence from 23 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Verified leaderboard positions: MiniMax M3 #18; Step 3.7 Flash unranked
BenchAlign evidence: MiniMax M3 supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. MiniMax M3 and Step 3.7 Flash share 23 comparable benchmark results. 2 of 8 categories are comparable. 22 results are unique to MiniMax M3; 7 to Step 3.7 Flash.
Updated July 16, 2026- Shared results
- 23
- MiniMax M3 only
- 22
- Step 3.7 Flash only
- 7
- Comparable categories
- 2 / 8
Pick MiniMax M3 if you want the stronger benchmark profile. Step 3.7 Flash 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 23 shared benchmark results across 6 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 57. 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 coding, where it averages 72.2 against 56.3. The single biggest benchmark swing on the page is BrowseComp, 83.5% to 75.8%.
MiniMax M3 is also the more expensive model on tokens at $0.30 input / $1.20 output per 1M tokens, versus $0.20 input / $1.15 output per 1M tokens for Step 3.7 Flash. Step 3.7 Flash 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 Step 3.7 Flash.
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 | MiniMax M3 | Δ | Step 3.7 Flash |
|---|---|---|---|
| Coding | MiniMax M372.2 | Margin← 15.9 | Step 3.7 Flash56.3 |
| Agentic | MiniMax M372.3 | Margin← 5.9 | Step 3.7 Flash66.4 |
| Math | MiniMax M385.7 | MarginNo overlap | Step 3.7 FlashNot measured |
| Multimodal | MiniMax M364.9 | MarginNo overlap | Step 3.7 FlashNot measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
BrowseComp
AgenticA 83.5%B 75.8%Winner: MiniMax M3Δ 7.7BrowseComp: MiniMax M3 scored 83.5%; Step 3.7 Flash scored 75.8%. MiniMax M3 wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 66%B 59.5%Winner: MiniMax M3Δ 6.5Terminal-Bench 2.0: MiniMax M3 scored 66%; Step 3.7 Flash scored 59.5%. MiniMax M3 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 59%B 56.3%Winner: MiniMax M3Δ 2.7SWE-bench Pro: MiniMax M3 scored 59%; Step 3.7 Flash scored 56.3%. MiniMax M3 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | MiniMax M3 | Step 3.7 Flash | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | MiniMax M3$0.3 input / $1.2 output | Step 3.7 Flash$0.2 input / $1.15 output | Step 3.7 Flash has the lower combined listed price. |
| Generation speedtokens per second | MiniMax M3Not available | Step 3.7 FlashNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | MiniMax M3Not available | Step 3.7 FlashNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | MiniMax M31M | Step 3.7 Flash256K | MiniMax M3 lists the larger context window. |
Benchmark Deep Dive
AgenticMiniMax M3 wins21 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 66% | 59.5% | MiniMax M3 leads |
| BrowseCompSource | 83.5% | 75.8% | MiniMax M3 leads |
| OSWorld-VerifiedSource | 70.1% | — | Not comparable |
| MCP AtlasSource | 74.2% | — | Not comparable |
| Claw-EvalSource | 74.5% | 67.1% | MiniMax M3 leads |
| AA Agentic IndexSource | 35.4% | 21.5% | MiniMax M3 leads |
| τ²-bench resultsSource | 88.9% | 98.5% | Step 3.7 Flash leads |
| GDPval-AASource | 44.7% | 25.9% | MiniMax M3 leads |
| GDPval-AASource | 1395 | 1017 | MiniMax M3 leads |
| 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 |
| DeepSearchQASource | — | 92.8% | Not comparable |
| ToolathlonSource | — | 49.5% | Not comparable |
| HLE w/ toolsSource | — | 47.2% | Not comparable |
| Gert LabsSource | — | 51.57% | Not comparable |
| APEX-Agents-AASource | — | 14.8% | Not comparable |
CodingMiniMax M3 wins11 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 80.5% | — | Not comparable |
| SWE-bench ProSource | 59% | 56.3% | MiniMax M3 leads |
| Terminal-Bench 2.0Source | 66.0% | 59.5% | MiniMax M3 leads |
| NL2RepoSource | 42.1% | — | Not comparable |
| AA Coding IndexSource | 58.6% | 39.6% | MiniMax M3 leads |
| Terminal-Bench HardSource | 42.4% | 35.6% | MiniMax M3 leads |
| AA-SciCodeSource | 45.4% | 40.0% | MiniMax M3 leads |
| 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
Knowledge7 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 44.4% | 30.3% | MiniMax M3 leads |
| AA-GPQA DiamondSource | 92.9% | 80.9% | MiniMax M3 leads |
| AA-HLESource | 37.1% | 19.9% | MiniMax M3 leads |
| AA-Omniscience IndexSource | 1.4% | -37.5% | MiniMax M3 leads |
| AA-Omniscience AccuracySource | 15.0% | 25.4% | Step 3.7 Flash leads |
| AA-Omniscience Hallucination RateSource | 16.1% | 84.4% | MiniMax M3 leads |
| AA Openness IndexSource | 33.3% | — | Not comparable |
Math1 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | Result |
|---|---|---|---|
| USAMO 2026Source | 85.7% | — | Not comparable |
Multimodal9 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | 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 | 1218 | MiniMax M3 leads |
| AA-MMMU-ProSource | 78.6% | 75.3% | MiniMax M3 leads |
| SimpleVQASource | — | 79.2% | Not comparable |
| V*Source | — | 95.3% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | MiniMax M3 | Step 3.7 Flash | Result |
|---|---|---|---|
| AA-IFBenchSource | 82.9% | 67.3% | MiniMax M3 leads |
Frequently Asked Questions (3)
Which is better, MiniMax M3 or Step 3.7 Flash?
MiniMax M3 is ahead on BenchLM's provisional leaderboard, 70 to 57. The biggest single separator in this matchup is BrowseComp, where the scores are 83.5% and 75.8%.
Which is better for coding, MiniMax M3 or Step 3.7 Flash?
MiniMax M3 has the edge for coding in this comparison, averaging 72.2 versus 56.3. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, MiniMax M3 or Step 3.7 Flash?
MiniMax M3 has the edge for agentic tasks in this comparison, averaging 72.3 versus 66.4. 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.