Model Timeline

Timeline of DeepSeek-V3.1-Think

Every recorded change for this model by DeepSeek, keyed on when it occurred. Back to the entity →

15 events

Filters

Year

Category (lane)Event typeFilters the loaded events (the API has no type filter on /timeline).OrganizationFamilyOrganization and family scope the timeline to that entity (one at a time).Min importanceInclude historical backfill

Reset

Lanes · dots sized by importance · drag to select a rangeoccurred vs observed vs recorded

ModelsBenchmarksProviders12 Sept 2026 → 12 Sept 2026

15 events loaded · every dot is one event; hover for its summary

Sept 2026 15digest →

12 Sept 2026
Listed by providerModelDeepSeek V3.1DeepSeek
OpenRouter lists DeepSeek V3.1 at $0.25 in / $0.95 out per 1M tokens
openrouter
openrouter.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 25% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 37.43% on τ²-bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 41.5% on IFBench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 14.27% on Humanity's Last Exam
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 77.88% on GPQA Diamond
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 13.48 on Artificial Analysis Intelligence Index
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 24.24% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 34.8% on τ²-bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 37.82% on IFBench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 6.67% on Humanity's Last Exam
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 73.54% on GPQA Diamond
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelDeepSeek V3.1DeepSeek
DeepSeek V3.1 scores 13.71 on Artificial Analysis Intelligence Index
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Property changedModelDeepSeek V3.1DeepSeek
DeepSeek V3.1: reasoning changed from false to true
ReasoningNo→Yesopenrouter
openrouter.ai
12 Sept 2026
Context window changedModelDeepSeek V3.1DeepSeek
DeepSeek V3.1: context length changed from 128000 to 163840
Context window128K tokens→163.8K tokensopenrouter
openrouter.ai

Showing up to 1,000 events. Dates are occurred_at (effective date when a source states one, else observation) — hover a date for the observation time. Historical backfill is excluded; the checkbox in the rail includes it. Cursor-paged history: changes feed.