Across 4 comparable benchmarks, MiniMax M3.1 Flash has better reported results on 4 and worse on 0. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence4 of 5 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Mercury 3 | MiniMax M3.1 Flash |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input / 1M | $0.100 | $0.150 |
| Output / 1M | $0.350 | $0.600 |
| Blended (4:1) | $0.085 | $0.143 |
| Status | Available | Available |
| Released | October 19, 2026 | October 19, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 2 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
MiniMax M3.1 Flash ahead on 1 of 1 counted
MiniMax M3.1 Flash ahead on 2 of 2 counted
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show Mercury 3 relative to MiniMax M3.1 Flash. Reading the results
| Benchmark | Mercury 3 | MiniMax M3.1 Flash | Gap |
|---|---|---|---|
| Terminal-Bench 4.0Agentic Development · Resolution rate (%)Not counted · Different or unrecorded test environments — excluded from the tally | 3.0Unspecified Reasoning · tools· vals-mini-swe | 38.1Unspecified Reasoning · tools· legacy-default | —#70 vs #30 of 90 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 60.9Thinking High · tools | 73.9≠Max · tools | −13.0 pts#59 vs #46 of 121 |
| Legal Research BenchLegal · Score | 28.9Thinking High · tools | 34.1≠Max · tools | 0.85×#58 vs #51 of 91 |
| Tax Agent BenchFinance · Accuracy (%, rubric × citation quality) | 18.2Thinking High · tools | 32.1≠Unspecified Reasoning · tools | −14.0 pts#56 vs #35 of 74 |
| Vals Index v2.2Capability Indices · GDP-weighted accuracy (%) | 43.7Thinking High · tools | 53.8≠Max · task tools | −10.1 pts#32 vs #24 of 36 |
Agentic Development · Resolution rate (%)
Not counted · Different or unrecorded test environments — excluded from the tally
Code Generation & Editing · Pass rate (%)
Legal · Score
Finance · Accuracy (%, rubric × citation quality)
Capability Indices · GDP-weighted accuracy (%)
A dash means no result is available.
| Benchmark | Mercury 3 | MiniMax M3.1 Flash |
|---|---|---|
| HLE | — | 48.0 |
| DeepSWE | — | 73.7 |
| FrontierCS 2 | — | 3.3 |
| NL2Repo-Bench | — | 61.2 |
| PostTrain Bench | — | 39.3 |
| SWE Refactor Bench | — | 23.0 |
| SWE-sweep | — | 0.67 |
| Agents' Last Exam v1.0 | — | 29.6 |
| Agents' Last Exam v1.1 | — | 29.5 |
| AutomationBench-AA | — | 72.5 |
| OSWorld 2.0 | — | 61.7 |
| 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 |
| 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 |
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%
Mercury 3 costs40%lessthan MiniMax M3.1 Flash per million tokens.Using standard rates at this token mix.
Mercury 3: Standard pricing. Inception Labs official docs/get-started/models.md verified 2026-11-13 (50% promo pricing prior to 2026-12-28 wind-down); Mercury 3 with 50% cache read promo
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.
| Price / 1M tokens | Mercury 3 | MiniMax M3.1 Flash | Ratio |
|---|---|---|---|
| Input | $0.10 | $0.15 | 0.67× |
| Output | $0.35 | $0.60 | 0.58× |
| Cached input | $0.01 | $0.015 | 0.67× |
| Cache write | — | — | — |
| Blended | $0.085 | $0.143 | 0.60× |
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.
Mercury 3 relative to MiniMax M3.1 Flash
Summary: 1 of 5 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 1 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Mercury 3 1.7× cheaper
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Mercury 3 1.7× cheaper
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ or are not recorded; interpret the comparison with care.
Mercury 3
MiniMax M3.1 Flash
Settings or sources differ or are not recorded; interpret the comparison with care.
Mercury 3
MiniMax M3.1 Flash
Effort, protocol, aggregation and source match; other run conditions may differ.
Mercury 3
MiniMax M3.1 Flash
Settings or sources differ or are not recorded; interpret the comparison with care.
Mercury 3
MiniMax M3.1 Flash
Settings or sources differ or are not recorded; interpret the comparison with care.
Mercury 3
MiniMax M3.1 Flash
| Benchmark | Mercury 3 | MiniMax M3.1 Flash |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Agents' Last Exam v1.0 | Not reported | |
| AutomationBench-AA | Not reported | $0.0665 / task Average · Guardrail-safe objective completion rate: 72.5 Thinking (With Tools) |
| FrontierCS 2 | Not reported | |
| GDPx | Not reported | $1.61 / task Average · GDPx — Score: 4.0 Highest (source-reported) (With Tools) · overall/score |
| HLE | Not reported | |
| LiveBench 2.0 | Not reported | $0.1056 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 73.1 Max (task-specific tools) |
| SWE-sweep | Not reported |
| Attribute | Mercury 3 | MiniMax M3.1 Flash |
|---|---|---|
| Availability | Available | Available |
| Provider | Inception Labs | MiniMaxAI |
| Model type | Reasoning model | Coding model |
| First release | October 19, 2026 | October 19, 2026 |
| Context window | 1M tokens | 1M tokens |
| Max output | 128K tokens | — |
| Total parameters | — | 428B |
| Active parameters | — | 23B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Tool use, Function calling, Structured output, Long context, Coding | Reasoning, Coding |
| Input modalities | Text | Text, Image, Video |
| Output modalities | Text | Text |
| Weights | Closed / API only | Open weights |
| 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 Mercury 3 or MiniMax M3.1 Flash? 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.