Across 10 comparable benchmarks, GPT-6 Sol has better reported results on 8 and worse on 2. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence10 of 12 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | GPT-6 Sol | GPT-5.1 |
|---|---|---|
| Context window | 1.05M tokens | 400K tokens |
| Input / 1M | $2.00 | $1.25 |
| Output / 1M | $10.00 | $10.00 |
| Blended (4:1) | $2.30 | $2.19 |
| Status | Available | Deprecated |
| Released | September 22, 2026 | November 12, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates β set your own mix.
Mean of 7 shared benchmarks grouped by capability on a 0β100 scale. How averages are built
GPT-6 Sol ahead on 1 of 1 counted
GPT-6 Sol ahead on 2 of 2 counted
Split 2β2 of 4 counted
GPT-6 Sol ahead on 1 of 1 counted
GPT-6 Sol ahead on 1 of 1 counted
No shared comparable results for Mathematics, Computer & tool use, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show GPT-6 Sol relative to GPT-5.1. Reading the results
| Benchmark | GPT-6 Sol | GPT-5.1 | Gap |
|---|---|---|---|
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 87.8Max Β· tools | 24.6β Thinking High Β· tools | +63.2 pts#14 vs #81 of 119 |
| ARC-AGI-1Abstract Generalization Β· Score (%) | 95.5Max Β· no tools | 72.8β Thinking High Β· no tools | +22.7 pts#18 vs #50 of 86 |
| ARC-AGI-2Abstract Generalization Β· Score (% solved); cost per task (USD)Not counted Β· Speed and cost telemetry is reported, not scored | 89.6Max Β· no tools | 17.6β Thinking High Β· no tools | 5.08Γ#11 vs #54 of 79 |
| SimpleBenchCommonsense Β· Score (AVG@5) | 73.1Historical | 53.2β Thinking High Β· no tools | 1.37Γ#21 vs #55 of 102 |
| Vending-Bench 2Logical Planning Β· Money Balance (USD)Not counted Β· Speed and cost telemetry is reported, not scored | 14428Unspecified Reasoning Β· tools | 1473Unspecified Reasoning Β· tools | 9.79Γ#5 vs #51 of 64 |
| CritPtScientific Reasoning Β· Score | 30.9Max Β· no tools | 4.9β Thinking High Β· no tools | 6.30Γ#9 vs #71 of 134 |
| GPQA DiamondScientific Reasoning Β· Accuracy | 94.3Max Β· no tools | 88.1β Thinking Β· no tools | +6.2 pts#13 vs #65 of 200 |
| MedCodeClinical Workflows Β· Accuracy (%) | 47.1Max Β· tools | 52.7β Thinking High Β· tools | β5.7 pts#35 vs #13 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 82.0Max Β· tools | 88.1β Thinking High Β· tools | β6.1 pts#48 vs #14 of 99 |
| MMMU-ProVisual Understanding Β· Accuracy | 82.9Max Β· no tools | 79.0β Thinking Β· no tools | +3.9 pts#15 vs #38 of 134 |
| SAGEService Workflows Β· Accuracy (%) | 44.8Max Β· tools | 43.2β Thinking High Β· tools | +1.6 pts#43 vs #51 of 82 |
| LiveBenchCross-capability Suites Β· Accuracy | 79.3Max Β· no tools | 72.0β Thinking High Β· no tools | +7.2 pts#12 vs #50 of 106 |
Code Generation & Editing Β· Pass rate (%)
Abstract Generalization Β· Score (%)
Abstract Generalization Β· Score (% solved); cost per task (USD)
Not counted Β· Speed and cost telemetry is reported, not scored
Commonsense Β· Score (AVG@5)
Logical Planning Β· Money Balance (USD)
Not counted Β· Speed and cost telemetry is reported, not scored
Scientific Reasoning Β· Score
Scientific Reasoning Β· Accuracy
Clinical Workflows Β· Accuracy (%)
Clinical Workflows Β· Rubric pass rate (%)
Visual Understanding Β· Accuracy
Service Workflows Β· Accuracy (%)
Cross-capability Suites Β· Accuracy
A dash means no result is available.
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 costs5.2%morethan GPT-5.1 per million tokens.Using listed rates vs standard 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
GPT-5.1: Standard pricing. OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-11-13; standard tier; Deprecated; shutdown 2027-04-01 (replacement: gpt-6-sol)
| Price / 1M tokens | GPT-6 Sol | GPT-5.1 | Ratio |
|---|---|---|---|
| Input | $2.00 | $1.25 | 1.60Γ |
| Output | $10.00 | $10.00 | 1.00Γ |
| Cached input | $0.20 | $0.125 | 1.60Γ |
| Cache write | $2.50 | β | β |
| Blended | $2.30 | $2.19 | 1.05Γ |
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 GPT-5.1
Summary: 2 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 2 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Effort, protocol, aggregation and source match; other run conditions may differ.
GPT-6 Sol
GPT-5.1
Effort, protocol, aggregation and source match; other run conditions may differ.
GPT-6 Sol
GPT-5.1
| Benchmark | GPT-6 Sol | GPT-5.1 |
|---|---|---|
| AA Coding Agent Index v1.5 | Not reported | |
| AA Intelligence Index v5.0.1 | 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 |
| 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 |
| CritPt | 2 more settings | Not reported |
| CWE-bench v1 | 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 |
| LiveBench 2.0 | $0.8654 / task Average Β· Overall score (equal-weight mean of 7 suite scores, 0-100): 81.9 Max (task-specific tools) | Not reported |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| MysteryMechanism | Not reported | |
| SAGE | Not reported | |
| Terminal-Bench 4.0 | $5.79 / task Average Β· Resolution rate (%): 44.4 Unspecified Reasoning (With Tools) Β· vals-mini-swe | Not reported |
| Terminal-Bench 4.1 | Not reported | |
| Vals Index v2.1 | Not reported | |
| 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 | GPT-5.1 |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | OpenAI | OpenAI |
| Model type | Reasoning model | Reasoning model |
| First release | September 22, 2026 | November 12, 2025 |
| Context window | 1.05M tokens | 400K tokens |
| Max output | 128K tokens | 128K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Coding, Function calling, Tool use, Structured output, Computer use, Web search, Long context | Reasoning, Multilingual |
| Knowledge cutoff | 2026-04-20 | β |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| 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 GPT-6 Sol or GPT-5.1? 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.