BenchLeader
Nex AGIOpen weightsReasoning modelAuto-detected

Nex-N2-Pro

Not yet ranked: too few independent results so far. Released 2 Jun 2026.

Blended price
$1.00/M
$0.500 in · $2.50 out
Output speed
116 tok/s
First answer
19 s
Context
262k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Knowledge51
  2. Instruction following65
  3. Long context67
  4. Composite65

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
GPQA Diamond (AA)not in index89.2%#80Artificial Analysis
Humanity's Last Exam (AA)not in index33.7%#86Artificial Analysis

Coding

BenchmarkdefaultSource
SciCode (AA)not in index43.3%#100Artificial Analysis

Agents & tools

BenchmarkdefaultSource
τ²-Bench Telecom (AA)not in index81.6%#112Artificial Analysis

Knowledge

BenchmarkdefaultSource
AA-Omniscience-27.1#207Artificial Analysis

Instruction following

BenchmarkdefaultSource
IFBench66.2%#97Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR81.7%#30Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index28.2#92Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.001021.5 s
Summarise a 30-page report12,000 / 600$0.007524.1 s
Code edit6,000 / 1,500$0.006831.8 s
Agentic coding session60,000 / 4,000$0.04053.3 s
Structured extraction2,000 / 200$0.001520.7 s

See also

Data as of 9 Sept 2026. Compare with another model.