Skip to content
Model Pareto

Nex-N2.5-Mini

Nex AGIOpen weightshuggingface.co ↗
Compare with nearest rivalsCompare with…
Input
—
Output
—
Speed
~120tok/sest.
Context
262Ktokens

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.

Hardware
1× H200 141GB
FP8 weights
Throughput
~8,390 tok/s
many requests batched
Price
$0.046–0.38 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, 300+ tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (35B params, 3B active per token) + 25% KV-cache headroom ≈ 44 GB
  • 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
  • ~8,390 output tok/s aggregate at batch 256 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~155,846 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

SWE-bench VerifiedEstimated
~#11 of 14 (estimated)~72.4%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. #10Claude Haiku 4.573.3%
  4. ~#11Nex-N2.5-MiniEstimated72.4%
  5. #11Nemotron 3 Ultra70.7%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Estimated
~#26 of 84 (estimated)~11.5%
  1. #1Claude Sonnet 5.563.6%
  2. #2Claude Opus 5.559.6%
  3. #23Kimi K312.6%
  4. ~#26Nex-N2.5-MiniEstimated11.5%
  5. #26Qwen3.8 2.4T A95B11.1%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#6 of 20 (estimated)~85.6%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #5gpt-oss-120b87.8%
  4. ~#6Nex-N2.5-MiniEstimated85.6%
  5. #6Gemma 4 31B80.0%
No published score — estimated from related benchmarks
SciCodeEstimated
~#28 of 96 (estimated)~50.8%
  1. #1Claude Opus 5.566.9%
  2. #2Claude Fable 5.163.1%
  3. #27DeepSeek V4 Pro (0813)51.0%
  4. ~#28Nex-N2.5-MiniEstimated50.8%
  5. #28Qwen3.8 Flash-Next50.6%
No published score — estimated from related benchmarks

Agents

OSWorld 2.0Estimated
~#4 of 5 (estimated)~48.5%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #3GPT-5.6 Terra50.2%
  4. ~#4Nex-N2.5-MiniEstimated48.5%
  5. #4GPT-5.547.5%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#26 of 87 (estimated)~1336 Elo
  1. #1Claude Opus 5.51846
  2. #2Claude Fable 5.11735
  3. #25GPT-5.51336
  4. ~#26Nex-N2.5-MiniEstimated1336
  5. #26MiniMax-M31230
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#34 of 80 (estimated)~23.4%
  1. #1Claude Opus 5.569.5%
  2. #2DeepSeek V4.1 Flash68.9%
  3. #33Quasar 438B23.9%
  4. ~#34Nex-N2.5-MiniEstimated23.4%
  5. #34MiniMax-M321.3%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#32 of 89 (estimated)~22.5%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #31Hy322.9%
  4. ~#32Nex-N2.5-MiniEstimated22.5%
  5. #32G9v3-39A5B22.1%
No published score — estimated from related benchmarks
Not comparable across labs (1)

Kept for reference, not counted in rankings: these use a lab-specific task set, answer key or scoring, so scores can't be compared fairly across labs. Only published scores are shown.