nex-agi/nex-n2.5-pro
Nex AGIvia OpenRouter- average±5.6 (95%)
- 53.2
- leaderboard rankof 114
- ≈101
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $2.78
- seconds per inputmean response time
- 498.9
By task
-
Data correction
56.1
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
69.1
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
59.3
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
39.5
Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.
By benchmark
| Benchmark | ScoreThe best current run on each benchmark, on the benchmark's metric scaled to 0–100. ± is the 95% sampling margin, from resampling the run's inputs. | RankRank among all current models on that benchmark; ≈ for models that cannot be told apart within sampling error. | Per 1,000 docs | s / input | Best run |
|---|---|---|---|---|---|
| Bibliographic Dataother prompt | 27.7±32.0 | ≈110of 115 | $2.63$0.013 run | 405.5 | T2208 · 25 Oct 2026prompt_v2.txt |
| Blacklist Cards | 86.7±3.0 | ≈79of 114 | $0.76$0.025 run | 85.67 | T2215 · 25 Oct 2026prompt.txt |
| Book Advert XML Repairother prompt | 32.3±9.2 | ≈41of 119 | $2.68$0.13 run | 409.53 | T2217 · 25 Oct 2026prompt_v2.txt |
| Business Letters | 56.0±7.2 | ≈75of 114 | $0.99$0.056 run | 133.98 | T2209 · 25 Oct 2026prompt.txt |
| Company Listsother prompt | 63.9±18.7 | ≈97of 114 | $2.15$0.032 run | 341.43 | T2216 · 25 Oct 2026prompt_v2.txt |
| Duty Rosters | 54.4±18.1 | ≈87of 107 | $9.72$0.20 run | 1702.42 | T2221 · 26 Oct 2026prompt.txt |
| Fraktur Adverts | 19.4±38.1 | ≈97of 114 | $3.43$0.017 run | 749.5 | T2212 · 25 Oct 2026prompt_optimized.txt |
| General Meeting Minutes | 40.8±22.8 | ≈103of 108 | $4.01$0.036 run | 451.86 | T2220 · 26 Oct 2026prompt.txt |
| Library Cardsother prompt | 85.5±2.1 | ≈51of 114 | $0.57$0.15 run | 108.56 | T2213 · 25 Oct 2026prompt_year-string.txt |
| Magazine Pagesother prompt | 72.9±12.2 | ≈34of 115 | $1.42$0.065 run | 216.58 | T2219 · 26 Oct 2026prompt_explicit-json.txt |
| Medieval Manuscripts | 19.2±20.0 | ≈107of 114 | $3.56$0.043 run | 729.54 | T2214 · 25 Oct 2026prompt.txt |
| Personnel Cards | 79.8±13.4 | ≈94of 114 | $1.49$0.091 run | 652.09 | T2218 · 25 Oct 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 24 Oct 2026 | 0.075 | 0.25 | archived price page |
Every run
- T2221Duty Rosters26 Oct 202654.4$0.20
- T2220General Meeting Minutes26 Oct 202640.8$0.036
- T2219Magazine Pages26 Oct 202672.9$0.065
- T2218Personnel Cards25 Oct 202679.8$0.091
- T2217Book Advert XML Repair25 Oct 202632.3$0.13
- T2216Company Lists25 Oct 202663.9$0.032
- T2215Blacklist Cards25 Oct 202686.7$0.025
- T2214Medieval Manuscripts25 Oct 202619.2$0.043
- T2213Library Cards25 Oct 202685.5$0.15
- T2212Fraktur Adverts25 Oct 202619.4$0.017
- T2211Business Letters25 Oct 2026 · 32 signed letters59.0$0.032
- T2210Business Letters25 Oct 2026 · 25 unsigned letters54.0$0.020
- T2209Business Letters25 Oct 2026 · all 57 letters56.0$0.056
- T2208Bibliographic Data25 Oct 202627.7$0.013