Across 26 comparable benchmarks, GPT-6 Sol has better reported results on 14 and worse on 11, with 1 tied. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence26 of 28 shared results countedHow we compare
Price basis: Listed rates vs international rates. The price section below shows the applicable periods.
| Fact | GPT-6 Sol | MiMo-V2.6-Pro |
|---|---|---|
| Context window | 1.05M tokens | 1M tokens |
| Input / 1M | $2.00 | $0.435 |
| Output / 1M | $10.00 | $0.870 |
| Blended (4:1) | $2.30 | $0.211 |
| Status | Available | Available |
| Released | September 22, 2026 | September 22, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates β set your own mix.
Mean of 17 shared benchmarks grouped by capability on a 0β100 scale. How averages are built
GPT-6 Sol ahead on 5 of 9 counted
GPT-6 Sol ahead on 1 of 1 counted
GPT-6 Sol ahead on 3 of 5 counted
MiMo-V2.6-Pro ahead on 1 of 1 counted
No fixed-scale average β see the rows in Benchmark results
MiMo-V2.6-Pro ahead on 4 of 6 counted
Tied on 1 counted
No shared comparable results for Abstract reasoning, Research & search, Long context, Multimodal.
Published scores, with matching modes where available. Gaps show GPT-6 Sol relative to MiMo-V2.6-Pro. Reading the results
| Benchmark | GPT-6 Sol | MiMo-V2.6-Pro | Gap |
|---|---|---|---|
| AA Cyber IndexRepository Engineering Β· Composite index (0β100) | 36.8Max Β· tools | 56.1β Thinking Β· tools | 0.66Γ#12 vs #4 of 22 |
| APEX-SWERepository Engineering Β· Pass@1 | 45.0Max Β· tools | 54.2β Auto Β· tools | β9.2 pts#26 vs #8 of 61 |
| Code MigrationMaintenance & Optimization Β· Hidden-test pass rate (%) | 57.2Max Β· tools | 43.0β Unspecified Reasoning Β· tools | +14.2 pts#10 vs #28 of 79 |
| DeepSWERepository Engineering Β· Pass@1 (DeepSWE v1.1) | 68.8Max Β· tools | 71.9β Thinking Β· tools | β3.1 pts#29 vs #21 of 73 |
| IOI (Vals v2)Algorithmic Coding Β· Subtask points (%) | 82.6Max Β· tools | 39.3β Unspecified Reasoning Β· tools | +43.3 pts#10 vs #36 of 45 |
| Terminal-Bench 2.1Agentic Development Β· Accuracy | 83.2Max Β· tools | 89.9β Thinking Β· tools | β6.8 pts#32 vs #6 of 68 |
| Terminal-Bench 4.0Agentic Development Β· Resolution rate (%) | 49.4Max Β· tools | 34.9β Thinking Β· tools | +14.5 pts#19 vs #34 of 90 |
| Terminal-Bench 4.1Agentic Development Β· Resolution rate (%) | 50.3Max Β· tools | 27.0β Normal Β· tools | +23.3 pts#15 vs #27 of 48 |
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 87.8Max Β· tools | 85.2β Unspecified Reasoning Β· tools | +2.6 pts#14 vs #20 of 121 |
| ProofBench v1.1 (Lean 4)Mathematics Β· Accuracy (%) | 83.0Max Β· tools | 70.0β Unspecified Reasoning Β· tools | +13.0 pts#17 vs #22 of 54 |
| CritPtScientific Reasoning Β· Score | 30.9Max Β· no tools | 26.6β Thinking Β· no tools | 1.16Γ#9 vs #19 of 134 |
| MysteryMechanismScientific Reasoning Β· Accuracy (%) | 30.2Max Β· tools | 15.3β Unspecified Reasoning Β· tools | +14.9 pts#17 vs #30 of 32 |
| SciCodeScientific Computing Β· Score | 57.6Max Β· no tools | 60.9β Thinking Β· no tools | 0.95Γ#22 vs #8 of 98 |
| MedCodeClinical Workflows Β· Accuracy (%) | 47.1Max Β· tools | 45.0β Unspecified Reasoning Β· tools | +2.1 pts#35 vs #37 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 82.0Max Β· tools | 88.3β Unspecified Reasoning Β· tools | β6.3 pts#48 vs #13 of 99 |
| APEX-AgentsTool Orchestration Β· Score | 54.3Max Β· tools | 54.4β Reported best | 1.00Γ#31 vs #30 of 61 |
| AutomationBenchOffice & Business Β· Pass RateNot counted Β· The test environment is not recorded β excluded from the tally | 33.2Extra-High Β· tools | 53.1β Thinking Β· tools | β#29 vs #7 of 64 |
| EMB (Excel Modeling)Finance Β· Accuracy (%) | 71.5Max Β· tools | 62.9β Unspecified Reasoning Β· tools | +8.7 pts#14 vs #32 of 78 |
| Finance Agent v2Finance Β· Score | 49.0Max Β· tools | 57.3β Unspecified Reasoning Β· tools | 0.86Γ#46 vs #14 of 75 |
| Legal Research BenchLegal Β· Score | 28.9Max Β· tools | 47.1β Unspecified Reasoning Β· tools | 0.61Γ#58 vs #16 of 91 |
| Public Benefits Bench v1.1Service Workflows Β· Criterion pass rate (%) | 56.6Max Β· tools | 68.9β Unspecified Reasoning Β· tools | β12.3 pts#39 vs #8 of 50 |
| SAGEService Workflows Β· Accuracy (%) | 44.8Max Β· tools | 45.0β Unspecified Reasoning Β· tools | β0.3 pts#43 vs #41 of 82 |
| Tax Agent BenchFinance Β· Accuracy (%, rubric Γ citation quality) | 53.0Max Β· tools | 32.4β Unspecified Reasoning Β· tools | +20.7 pts#23 vs #34 of 74 |
| ExploitGymExploitation Β· Score (%)Not counted Β· Different or unrecorded test environments β excluded from the tally | 22.1Reported bestΒ· exploitgym-unlimited | 17.8β Thinking Β· toolsΒ· exploitgym-budget-unspecified | β#5 vs #7 of 12 |
| SEC-Bench ProVulnerability Analysis Β· Score (%) | 66.3Reported best | 66.3β Thinking Β· tools | =#8 vs #8 of 13 |
| AA Intelligence Index v5.0.1Capability Indices Β· Index (0-100) | 44.2Max Β· task tools | 29.9β Thinking Β· tools | 1.48Γ#16 vs #31 of 62 |
| LiveBench 2.0Cross-capability Suites Β· Overall score (equal-weight mean of 7 suite scores, 0-100) | 81.9Max Β· task tools | 74.9β Thinking High Β· tools | 1.09Γ#9 vs #19 of 51 |
| Vals Index v2.1Capability Indices Β· GDP-weighted accuracy (%) | 57.5Max Β· task tools | 55.2β Reported best | +2.3 pts#10 vs #12 of 38 |
Repository Engineering Β· Composite index (0β100)
Repository Engineering Β· Pass@1
Maintenance & Optimization Β· Hidden-test pass rate (%)
Repository Engineering Β· Pass@1 (DeepSWE v1.1)
Algorithmic Coding Β· Subtask points (%)
Agentic Development Β· Accuracy
Agentic Development Β· Resolution rate (%)
Agentic Development Β· Resolution rate (%)
Code Generation & Editing Β· Pass rate (%)
Mathematics Β· Accuracy (%)
Scientific Reasoning Β· Score
Scientific Reasoning Β· Accuracy (%)
Scientific Computing Β· Score
Clinical Workflows Β· Accuracy (%)
Clinical Workflows Β· Rubric pass rate (%)
Tool Orchestration Β· Score
Office & Business Β· Pass Rate
Not counted Β· The test environment is not recorded β excluded from the tally
Finance Β· Accuracy (%)
Finance Β· Score
Legal Β· Score
Service Workflows Β· Criterion pass rate (%)
Service Workflows Β· Accuracy (%)
Finance Β· Accuracy (%, rubric Γ citation quality)
Exploitation Β· Score (%)
Not counted Β· Different or unrecorded test environments β excluded from the tally
Vulnerability Analysis Β· Score (%)
Capability Indices Β· Index (0-100)
Cross-capability Suites Β· Overall score (equal-weight mean of 7 suite scores, 0-100)
Capability Indices Β· GDP-weighted accuracy (%)
A dash means no result is available.
| Benchmark | GPT-6 Sol | MiMo-V2.6-Pro |
|---|---|---|
| FrontierMath ErdΕs | 1.5 | β |
| FrontierMath Tier 4 v2 | 90.0 | β |
| FrontierMath v2 | 89.8 | β |
| GPQA Diamond | 94.3 | β |
| Riemann-bench | 64.8 | β |
| AA Coding Agent Index v1.5 | 56.7 | β |
| FrontierCode 1.1 | 49.3 | β |
| FrontierCode 2.0 | 34.3 | β |
| ProgramBench | β | 26.5 |
| WeirdML v3 | 0.2 | β |
| Agents' Last Exam v1.0 | 32.2 | β |
| Job Bench | β | 62.0 |
| OSWorld 2.0 | 64.4 | β |
| OSWorld-Verified | β | 82.0 |
| Toolathlon-Verified | β | 76.9 |
| CWE-bench v1 | 52.0 | β |
| CyberGym | β | 94.0 |
| Agent Arena | 6.8 | β |
| Agent Arena Index | 1178 | β |
| AgentTime v1.1 | β | 18.0 |
| LiveBench | 79.3 | β |
| Vals Index | 62.6 | β |
| Vals Index v2.2 | β | 57.8 |
| Creative Writing | 2126 | β |
| AA-LCR | β | 86.3 |
| EBR-bench | 53.3 | β |
| GDP.pdf | 25.2 | β |
| MMMU-Pro | 82.9 | β |
| GDPval-AA v2 | β | 1691 |
| Harvey's Legal Agent Benchmark | 1.7 | β |
| ARC-AGI-1 | 95.5 | β |
| ARC-AGI-2 | 89.6 | β |
| ARC-AGI-3 | 23.0 | β |
| GoBench | 2006 | β |
| SimpleBench | 73.1 | β |
| Vending-Bench 2 | 14428 | β |
| BioMysteryBench | 74.8 | β |
| HealthBench | 53.2 | β |
| HealthBench Professional | 60.8 | β |
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 Β· 5% steps
Blended price = effective input rate at 90% cache hits Γ 80% + output rate Γ 20%
GPT-6 Sol costs990%morethan MiMo-V2.6-Pro per million tokens β 10.9Γ the price.Using listed rates vs international rates at this token mix.
GPT-6 Sol: Standard pricing. context_length <= 272000 OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-10-21; standard tier; <=272K input tokens context_length > 272000 OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-10-21; standard tier; >272K input tokens
MiMo-V2.6-Pro: International pricing. Xiaomi MiMo international pricing β input Xiaomi MiMo international pricing β output
| Price / 1M tokens | GPT-6 Sol | MiMo-V2.6-Pro | Ratio |
|---|---|---|---|
| Input | $2.00 | $0.435 | 4.60Γ |
| Output | $10.00 | $0.87 | 11.5Γ |
| Cached input | $0.20 | $0.004 | 55.6Γ |
| Cache write | $2.50 | β | β |
| Blended | $2.30 | $0.211 | 10.9Γ |
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.
GPT-6 Sol relative to MiMo-V2.6-Pro
Summary: 1 of 7 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.
GPT-6 Sol 12.6Γ more expensive
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
GPT-6 Sol 12.6Γ more expensive
Settings or sources differ
Settings or sources differ
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.
GPT-6 Sol
MiMo-V2.6-Pro
Settings or sources differ or are not recorded; interpret the comparison with care.
GPT-6 Sol
MiMo-V2.6-Pro
Settings or sources differ or are not recorded; interpret the comparison with care.
GPT-6 Sol
MiMo-V2.6-Pro
Settings or sources differ or are not recorded; interpret the comparison with care.
GPT-6 Sol
MiMo-V2.6-Pro
Effort, protocol, aggregation and source match; other run conditions may differ.
GPT-6 Sol
MiMo-V2.6-Pro
Settings or sources differ or are not recorded; interpret the comparison with care.
GPT-6 Sol
MiMo-V2.6-Pro
Settings or sources differ or are not recorded; interpret the comparison with care.
GPT-6 Sol
MiMo-V2.6-Pro
| Benchmark | GPT-6 Sol | MiMo-V2.6-Pro |
|---|---|---|
| AA Coding Agent Index v1.5 | Not reported | |
| AA-LCR | Not reported | |
| Agent Arena | $1.8407 / task Median Β· Net improvement index (percentage points vs. average model): 6.8 Max (With Tools + Internet) | Not reported |
| Agent Arena Index | $1.872 / task Median Β· Agent Arena Index rating (index points): 1178 Max (With Tools + Internet) | Not reported |
| Agents' Last Exam v1.0 | 3 more settings | Not reported |
| ARC-AGI-1 | 3 more settings | Not reported |
| ARC-AGI-2 | 3 more settings | Not reported |
| AutomationBench | $0.24 / task Average Β· Pass Rate: 31.2 Thinking High (With Tools) Β· zapier-private-heldout 3 more settings$0.19 / task Average Β· Pass Rate: 21.2 Thinking Low (With Tools) Β· zapier-private-heldout $0.21 / task Average Β· Pass Rate: 26.9 Thinking Medium (With Tools) Β· zapier-private-heldout $0.27 / task Average Β· Pass Rate: 33.2 Extra-High (With Tools) Β· zapier-private-heldout | Not reported |
| Code Migration | Not reported | $0.6916 / task Average Β· Hidden-test pass rate (%): 43.0 Unspecified Reasoning (With Tools) |
| CWE-bench v1 | Not reported | |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| FrontierCode 2.0 | $1.0714 / task Average Β· Weighted rubric score: 22.8 Thinking High (With Tools + Internet) $0.4164 / task Average Β· Weighted rubric score: 10.5 Thinking Low (With Tools + Internet) 2 more settings$0.8225 / task Average Β· Weighted rubric score: 21.6 Thinking Medium (With Tools + Internet) | Not reported |
| GDPval-AA v2 | Not reported | |
| IOI (Vals v2) | Not reported | |
| Legal Research Bench | Not reported | |
| MedCode | Not reported | |
| MedScribe | Not reported | $0.00988 / task Average Β· Rubric pass rate (%): 88.3 Unspecified Reasoning (With Tools) |
| ProofBench v1.1 (Lean 4) | Not reported | |
| Public Benefits Bench v1.1 | Not reported | $0.0819 / task Average Β· Criterion pass rate (%): 68.9 Unspecified Reasoning (With Tools) |
| SAGE | Not reported | |
| SciCode | Not reported | |
| Tax Agent Bench | Not reported | $0.1343 / task Average Β· Accuracy (%, rubric Γ citation quality): 32.4 Unspecified Reasoning (With Tools) |
| Vals Index v2.2 | Not reported | $0.4347 / task Average Β· GDP-weighted accuracy (%): 57.8 Reported best (effort unspecified) |
| Vibe Code Bench v1.1 | Not reported | |
| WeirdML v3 | $6.9054 / task Average Β· Official score (0-1; higher is better): 0.2 Extra-High (With Tools) | Not reported |
| Attribute | GPT-6 Sol | MiMo-V2.6-Pro |
|---|---|---|
| Availability | Available | Available |
| Provider | OpenAI | Xiaomi |
| Model type | Reasoning model | Multimodal model |
| First release | September 22, 2026 | September 22, 2026 |
| Context window | 1.05M tokens | 1M tokens |
| Max output | 128K tokens | β |
| Total parameters | β | 1.0T |
| Active parameters | β | 42B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual, Coding, Function calling, Tool use, Structured output, Computer use, Web search, Long context | Reasoning, Multilingual, Coding, Tool use, Function calling, Computer use, Long context |
| Knowledge cutoff | 2026-04-20 | β |
| Input modalities | Text, Image | Text, Image, Audio, Video |
| Output modalities | Text | Text |
| Weights | Closed / API only | Open weights |
| Licence | β | MIT License |
| Commercial use | Proprietary | Commercial use allowed |
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 GPT-6 Sol or MiMo-V2.6-Pro? 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.