Across 2 comparable benchmarks, MiniMax M3.1 Flash and Qwen3-Max-Thinking are evenly matched at 1 apiece.
Limited evidence2 of 2 shared results countedHow we compare
Price basis: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | MiniMax M3.1 Flash | Qwen3-Max-Thinking |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input / 1M | $0.150 | — |
| Output / 1M | $0.600 | — |
| Blended (4:1) | $0.143 | — |
| Status | Available | Deprecated |
| Released | October 19, 2026 | January 26, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
MiniMax M3.1 Flash ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show MiniMax M3.1 Flash relative to Qwen3-Max-Thinking. Reading the results
| Benchmark | MiniMax M3.1 Flash | Qwen3-Max-Thinking | Gap |
|---|---|---|---|
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 73.9Max · tools | 3.5≠Unspecified Reasoning · tools | +70.4 pts#46 vs #110 of 121 |
| HLEKnowledge Exams · Accuracy | 48.0Reported best | 49.8≠Thinking · tools | −1.8 pts#58 vs #51 of 151 |
Code Generation & Editing · Pass rate (%)
Knowledge Exams · Accuracy
A dash means no result is available.
| Benchmark | MiniMax M3.1 Flash | Qwen3-Max-Thinking |
|---|---|---|
| C-Eval | — | 93.7 |
| CritPt | — | 1.7 |
| GPQA Diamond | — | 87.4 |
| HLE-Verified | — | 38.2 |
| IMO-AnswerBench | — | 83.9 |
| MMLU-Pro | — | 85.7 |
| DeepSWE | 73.7 | — |
| FrontierCS 2 | 3.3 | — |
| LiveCodeBench | — | 85.9 |
| NL2Repo-Bench | 61.2 | — |
| PostTrain Bench | 39.3 | — |
| SWE Refactor Bench | 23.0 | — |
| SWE-bench Verified | — | 75.3 |
| SWE-sweep | 0.67 | — |
| Terminal Bench Hard | — | 24.2 |
| Terminal-Bench 4.0 | 38.1 | — |
| Agents' Last Exam v1.0 | 29.6 | — |
| Agents' Last Exam v1.1 | 29.5 | — |
| AutomationBench-AA | 72.5 | — |
| OSWorld 2.0 | 61.7 | — |
| Pinch Bench | — | 80.3 |
| CyberGym | 82.2 | — |
| AA Intelligence Index v4.3 | 42.4 | — |
| AA Intelligence Index v5.0.1 | 29.1 | — |
| Code Arena WebDev | 1588 | — |
| LiveBench 2.0 | 73.1 | — |
| Vals Index v2.2 | 53.8 | — |
| IF Bench | — | 70.9 |
| LOCA-Bench 256K | 60.3 | — |
| BabyVision | 91.0 | — |
| CharXiv RQ | 82.5 | — |
| MMMU-Pro | 79.1 | — |
| AutomationBench | 62.8 | — |
| BankerToolBench | 76.2 | — |
| GDPx | 4.0 | — |
| Legal Research Bench | 34.1 | — |
| Tax Agent Bench | 32.1 | — |
| τ²-Bench | — | 82.1 |
| MedCode | — | 31.4 |
| MedScribe | — | 72.7 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
MiniMax M3.1 Flash: Standard pricing. MiniMax international platform (platform.minimax.io) pay-as-you-go list price, flat across context lengths. Official independent currency quote; no conversion.
Qwen3-Max-Thinking: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | MiniMax M3.1 Flash | Qwen3-Max-Thinking | Ratio |
|---|---|---|---|
| Input | $0.15 | — | — |
| Output | $0.60 | — | — |
| Cached input | $0.015 | — | — |
| Cache write | — | — | — |
| Blended | $0.143 | — | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
MiniMax M3.1 Flash relative to Qwen3-Max-Thinking
No matching recorded settings for a cost summary.
Summary: 0 of 1 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 0 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Settings or sources differ
Settings or sources differ
Settings or sources differ or are not recorded; interpret the comparison with care.
MiniMax M3.1 Flash
Qwen3-Max-Thinking
| Benchmark | MiniMax M3.1 Flash | Qwen3-Max-Thinking |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Agents' Last Exam v1.0 | Not reported | |
| AutomationBench-AA | $0.0665 / task Average · Guardrail-safe objective completion rate: 72.5 Thinking (With Tools) | Not reported |
| FrontierCS 2 | Not reported | |
| GDPx | $1.61 / task Average · GDPx — Score: 4.0 Highest (source-reported) (With Tools) · overall/score | Not reported |
| HLE | Not reported | |
| Legal Research Bench | Not reported | |
| LiveBench 2.0 | $0.1056 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 73.1 Max (task-specific tools) | Not reported |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| SWE-sweep | Not reported | |
| Tax Agent Bench | $0.3209 / task Average · Accuracy (%, rubric × citation quality): 32.1 Unspecified Reasoning (With Tools) | Not reported |
| Terminal-Bench 4.0 | $0.73 / task Average · Resolution rate (%): 25.3 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Vals Index v2.2 | Not reported |
| Attribute | MiniMax M3.1 Flash | Qwen3-Max-Thinking |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | MiniMaxAI | Alibaba |
| Model type | Coding model | Reasoning model |
| First release | October 19, 2026 | January 26, 2026 |
| Context window | 1M tokens | 1M tokens |
| Max output | — | 32K tokens |
| Total parameters | 428B | 1.0T |
| Active parameters | 23B | — |
| Architecture | Mixture of Experts (MoE) | Dense |
| Documented capabilities | Reasoning, Coding | Reasoning, Multilingual |
| Knowledge cutoff | — | 2024-10 |
| Input modalities | Text, Image, Video | Text |
| Output modalities | Text | Text |
| Weights | Open weights | Closed / API only |
| Licence | MiniMax M3.1-Flash Community License | — |
| Commercial use | Proprietary | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used MiniMax M3.1 Flash or Qwen3-Max-Thinking? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.