Skip to content
Model Pareto

Compare models

Side by side: quality per category, price, speed, and every benchmark cell with where it came from. Pick up to four.
Ling 3.0 FlashNex-N2.5-MiniSolar Pro 4
Focus

Coding: quality vs price

Compared models are ringed; the other 100 ranked here are greyed.

Lower price is better. Pareto frontier: Granite 4.2 3B, NVIDIA Nemotron 3 Nano 30B A3B, gpt-oss-20b, Ling-3.0-flash-VL, MiMo-V2.6-Flash, GLM-5.3 Flash, DeepSeek V4.1 Flash, MiMo-V2.6-Pro, Claude Sonnet 5.5, Claude Opus 5.5. 26 models are not plotted: Xing4.0-29B-A4B, Nex-N2.5-Pro, Atria Dawn Preview, Intern-S2-397B, Solar Open2 250B, Phi-4-reasoning-plus, Motif 3, Nex-N2-Pro, Nex-N2.5-Mini, A.X-K2, K2 Horizon 375B A23B, K-EXAONE 2.0 0803, K2 Horizon MoVA 36B A4B, Intern-S2-Preview (35B-A3B), Command A+, North Mini Code, G9v3-39A5B, Falcon-H1-34B-Instruct, Devstral 2, K2 Horizon 7B, Devstral Small 2, MiniCPM5-2B, Ling 3.0 Tiny, K2 Horizon 3.7B, K2 Horizon 0.9B, LFM2.5-2.6B.

26 models with no price data — shown in the strip at the left edge

Xing4.0-29B-A4B, Nex-N2.5-Pro, Atria Dawn Preview, Intern-S2-397B, Solar Open2 250B, Phi-4-reasoning-plus, Motif 3, Nex-N2-Pro, Nex-N2.5-Mini, A.X-K2, K2 Horizon 375B A23B, K-EXAONE 2.0 0803, K2 Horizon MoVA 36B A4B, Intern-S2-Preview (35B-A3B), Command A+, North Mini Code, G9v3-39A5B, Falcon-H1-34B-Instruct, Devstral 2, K2 Horizon 7B, Devstral Small 2, MiniCPM5-2B, Ling 3.0 Tiny, K2 Horizon 3.7B, K2 Horizon 0.9B, LFM2.5-2.6B

  • Best-value frontier (nothing is both cheaper and better)
  • Evidencestrong → weak
  • Estimated from other categories
  • Ling 3.0 Flash
    Quality
    34.3
    Rank
    #63/103
    Price
    $0.11/M tok
    Speed
    346tok/s
    In$0.075▾Out$0.22▾/M tok

    vs Solar Pro 4: −3.3 quality · 0.21× price

  • Nex-N2.5-Mini
    Quality
    36.7
    Rank
    #52/103
    Price
    —
    Speed
    —

    vs Solar Pro 4: −0.9 quality · price n/a

  • Solar Pro 4
    Quality
    37.5
    Rank
    #49/103
    Price
    $0.52/M tok
    Speed
    93tok/s
    In$0.30Out$1.20/M tok

    vs Nex-N2.5-Mini: +0.9 quality · price n/a

Overview

Metric
Ling 3.0 Flash
Nex-N2.5-Mini
Solar Pro 4
Lab
inclusionAI (Ant Group)
Nex AGI
Upstage
Released
Aug 4, 2026
Sep 8, 2026
Aug 6, 2026
Weights
Open
Open
Proprietary
Input price
$/M tok
$0.075Best
—
$0.30
Output price
$/M tok
$0.22Best
—
$1.20
Blended price
$/M tok · 3:1 in:out
$0.111Best
—
$0.525
Output speed
tok/s
346Best
—
93
Context
tokens
262K
262K
512KBest

Quality by category

0–100 within each category (100 = best tracked model). Rank is among all models ranked there. Click a row to focus it.

Metric
Ling 3.0 Flash
Nex-N2.5-Mini
Solar Pro 4
Coding
34.3±8.0
#63/103Verified
36.7±8.6
#52/103Lab-reported
37.5±8.0Best
#49/103Verified
Hard reasoning
58.7±5.8
#49/182Verifiedbest value
Not ranked
64.7±5.8Best
#38/182Verified
Agents
49.8±10.0Best
#34/173Verifiedbest value
49.1±8.9
#35/173Lab-reported
47.9±10.0
#36/173Verified

Coding benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Ling 3.0 Flash
Nex-N2.5-Mini
Solar Pro 4
SWE-bench Verified
day 0
~#10/14~73.9%Estimated
  1. #1Claude Opus 596.0%
  2. #9Xing4.0-29B-A4B75.0%
  3. ~#10Ling 3.0 FlashEstimated73.9%
  4. #10Claude Haiku 4.573.3%
estimated from related benchmarks
~#11/14~72.4%Estimated
  1. #1Claude Opus 596.0%
  2. #10Claude Haiku 4.573.3%
  3. ~#11Nex-N2.5-MiniEstimated72.4%
  4. #11Nemotron 3 Ultra70.7%
estimated from related benchmarks
~#9/14~75.4%Estimated
  1. #1Claude Opus 596.0%
  2. #8Muse Glimmer76.0%
  3. ~#9Solar Pro 4Estimated75.4%
  4. #9Xing4.0-29B-A4B75.0%
estimated from related benchmarks
SWE-bench Pro (Public)
day 0
~#17/19~53.0%Estimated
  1. #1Claude Opus 5.589.9%
  2. #15Gemini 3.5 Flash-Lite54.2%
  3. ~#17Ling 3.0 FlashEstimated53.0%
  4. #17Muse Glimmer51.2%
estimated from related benchmarks
#18/1843.8%Lab-reported
  1. #1Claude Opus 5.589.9%
  2. #17Muse Glimmer51.2%
  3. #18Nex-N2.5-Mini43.8%
github.com · 2026-09-08
~#14/19~54.6%Estimated
  1. #1Claude Opus 5.589.9%
  2. #13Hy357.9%
  3. ~#14Solar Pro 4Estimated54.6%
  4. #14Inkling54.3%
estimated from related benchmarks
DeepSWE v1.1
day 0
~#28/31~49.5%Estimated
  1. #1DeepSeek V4.1 Flash74.2%
  2. #27Qwen3.8 Max (0902)51.0%
  3. ~#28Ling 3.0 FlashEstimated49.5%
  4. #28Qwen3.8 27B42.2%
estimated from related benchmarks
#29/3036.1%Lab-reported
  1. #1DeepSeek V4.1 Flash74.2%
  2. #28Qwen3.8 27B42.2%
  3. #29Nex-N2.5-Mini36.1%
  4. #30Hy328.0%
github.com · 2026-09-08
~#27/31~51.5%Estimated
  1. #1DeepSeek V4.1 Flash74.2%
  2. #26Nex-N2.5-Pro55.8%
  3. ~#27Solar Pro 4Estimated51.5%
  4. #27Qwen3.8 Max (0902)51.0%
estimated from related benchmarks
Terminal-Bench 4.0
day 0
#49/830.0%Verified
  1. #1Claude Sonnet 5.563.6%
  2. #49K2 Horizon MoVA 36B A4B0.0%
  3. #49Ling 3.0 Flash0.0%
  4. #49Ling 3.0 Tiny0.0%
artificialanalysis.ai · 2026-09-24
~#26/84~11.5%Estimated
  1. #1Claude Sonnet 5.563.6%
  2. #23Kimi K312.6%
  3. ~#26Nex-N2.5-MiniEstimated11.5%
  4. #26Qwen3.8 2.4T A95B11.1%
estimated from related benchmarks
#39/830.5%BestVerified
  1. #1Claude Sonnet 5.563.6%
  2. #39Ring-2.6-1T0.5%
  3. #39Solar Pro 40.5%
  4. #39Trinity Large Thinking0.5%
artificialanalysis.ai · 2026-09-24
LiveCodeBench
third-party
~#8/20~70.6%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #7gpt-oss-20b77.7%
  3. ~#8Ling 3.0 FlashEstimated70.6%
  4. #8Mistral Small 464.0%
estimated from related benchmarks
~#6/20~85.6%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5gpt-oss-120b87.8%
  3. ~#6Nex-N2.5-MiniEstimated85.6%
  4. #6Gemma 4 31B80.0%
estimated from related benchmarks
~#8/20~75.1%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #7gpt-oss-20b77.7%
  3. ~#8Solar Pro 4Estimated75.1%
  4. #8Mistral Small 464.0%
estimated from related benchmarks
SciCode
third-party
#54/9542.0%Verified
  1. #1Claude Opus 5.566.9%
  2. #54K-EXAONE 2.0 080342.0%
  3. #54Ling 3.0 Flash42.0%
  4. #56Gemini 3.5 Flash-Lite41.3%
artificialanalysis.ai · 2026-09-24
~#28/96~50.8%Estimated
  1. #1Claude Opus 5.566.9%
  2. #27DeepSeek V4 Pro (0813)51.0%
  3. ~#28Nex-N2.5-MiniEstimated50.8%
  4. #28Qwen3.8 Flash-Next50.6%
estimated from related benchmarks
#45/9544.6%BestVerified
  1. #1Claude Opus 5.566.9%
  2. #44Qwen3.5 397B A17B44.8%
  3. #45Solar Pro 444.6%
  4. #46Ling-3.0-flash-VL44.2%
artificialanalysis.ai · 2026-09-24
AA Coding Agent Index
third-party
~last/15below measured range (<41.1)Estimated
  1. #1Claude Opus 5.566.0
  2. #14GPT-6 Luna41.1
  3. ~#15Ling 3.0 FlashEstimated<41.1
estimated from related benchmarks
~#13/15~42.8Estimated
  1. #1Claude Opus 5.566.0
  2. #12DeepSeek V4 Pro (0813)43.1
  3. ~#13Nex-N2.5-MiniEstimated42.8
  4. #13Gemini 3.8 Flash41.9
estimated from related benchmarks
~last/15below measured range (<41.1)Estimated
  1. #1Claude Opus 5.566.0
  2. #14GPT-6 Luna41.1
  3. ~#15Solar Pro 4Estimated<41.1
estimated from related benchmarks
Benchmarks in other categories (2)

Hard reasoning benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Ling 3.0 Flash
Nex-N2.5-Mini
Solar Pro 4
GPQA Diamond
day 0
#44/17285.5%Verified
  1. #1GPT-6 Astra96.1%
  2. #44KAT Coder Pro V285.5%
  3. #44Ling 3.0 Flash85.5%
  4. #46MiMo-V2.584.9%
artificialanalysis.ai · 2026-09-24
—
#32/17289.1%BestVerified
  1. #1GPT-6 Astra96.1%
  2. #31Nex-N2-Pro89.2%
  3. #32Solar Pro 489.1%
  4. #33GPT-5.4 mini87.5%
artificialanalysis.ai · 2026-09-24
Humanity's Last Exam
day 0
#51/17823.7%Verified
  1. #1Claude Opus 5.561.4%
  2. #50Qwen3.5 122B A10B25.2%
  3. #51Ling 3.0 Flash23.7%
  4. #52Gemma 4 31B23.6%
artificialanalysis.ai · 2026-09-24
—
#43/17829.2%BestVerified
  1. #1Claude Opus 5.561.4%
  2. #42A.X-K229.6%
  3. #43Solar Pro 429.2%
  4. #44Qwen3.5 397B A17B29.0%
artificialanalysis.ai · 2026-09-24
AIME (latest)
day 0
~#2/25~94.8%Estimated
  1. #1Inkling97.1%
  2. ~#2Ling 3.0 FlashEstimated94.8%
  3. #2Muse Glimmer94.7%
estimated from related benchmarks
—
~#2/25~94.7%Estimated
  1. #1Inkling97.1%
  2. ~#2Solar Pro 4Estimated94.7%
  3. #2Muse Glimmer94.7%
estimated from related benchmarks
AA Intelligence Index
third-party
~#38/75~24.4Estimated
  1. #1Claude Opus 5.557.6
  2. #37Ling-3.0-flash-VL24.6
  3. ~#38Ling 3.0 FlashEstimated24.4
  4. #38GPT-5.4 mini24.1
estimated from related benchmarks
—
~#29/75~27.9Estimated
  1. #1Claude Opus 5.557.6
  2. #28MiniMax-M329.2
  3. ~#29Solar Pro 4Estimated27.9
  4. #29Quasar 438B26.7
estimated from related benchmarks

Agents benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Ling 3.0 Flash
Nex-N2.5-Mini
Solar Pro 4
τ²-bench (Telecom)
day 0
~#37/116~75.5%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #36GPT-5.4 nano76.0%
  3. ~#37Ling 3.0 FlashEstimated75.5%
  4. #37Nova 2.0 Lite72.8%
estimated from related benchmarks
~#47/116~50.3%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #46Nemotron Cascade 2 30B A3B53.2%
  3. ~#47Nex-N2.5-MiniEstimated50.3%
  4. #47Solar Open 100B48.2%
estimated from related benchmarks
~#28/116~81.5%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #26Qwen3.5 2B81.6%
  3. ~#28Solar Pro 4Estimated81.5%
  4. #28MiniCPM5-1B81.0%
estimated from related benchmarks
OSWorld-Verified
day 0
~#9/12~71.5%Estimated
  1. #1Kimi K384.8%
  2. #8Gemini 3.5 Flash-Lite74.0%
  3. ~#9Ling 3.0 FlashEstimated71.5%
  4. #9Nex-N2.5-Mini71.2%
estimated from related benchmarks
#9/1171.2%Lab-reported
  1. #1Kimi K384.8%
  2. #8Gemini 3.5 Flash-Lite74.0%
  3. #9Nex-N2.5-Mini71.2%
  4. #10Muse Glimmer65.9%
github.com · 2026-09-08
~#10/12~70.5%Estimated
  1. #1Kimi K384.8%
  2. #9Nex-N2.5-Mini71.2%
  3. ~#10Solar Pro 4Estimated70.5%
  4. #10Muse Glimmer65.9%
estimated from related benchmarks
OSWorld 2.0
day 0
~last/5below measured range (<47.5%)Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #4GPT-5.547.5%
  3. ~#5Ling 3.0 FlashEstimated<47.5%
estimated from related benchmarks
~#4/5~48.5%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #3GPT-5.6 Terra50.2%
  3. ~#4Nex-N2.5-MiniEstimated48.5%
  4. #4GPT-5.547.5%
estimated from related benchmarks
~last/5below measured range (<47.5%)Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #4GPT-5.547.5%
  3. ~#5Solar Pro 4Estimated<47.5%
estimated from related benchmarks
BrowseComp
day 0
~#15/18~72.6%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #14Inkling77.1%
  3. ~#15Ling 3.0 FlashEstimated72.6%
  4. #15Mistral Medium 3.548.6%
estimated from related benchmarks
#13/1783.4%Lab-reported
  1. #1Atria Dawn Preview92.5%
  2. #12MiniMax-M383.5%
  3. #13Nex-N2.5-Mini83.4%
  4. #14Inkling77.1%
github.com · 2026-09-08
~#15/18~69.9%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #14Inkling77.1%
  3. ~#15Solar Pro 4Estimated69.9%
  4. #15Mistral Medium 3.548.6%
estimated from related benchmarks
GDPval-AA (Elo)
third-party
~#23/87~1386Estimated
  1. #1Claude Opus 5.51846
  2. #22Qwen3.8 27B1409
  3. ~#23Ling 3.0 FlashEstimated1386
  4. #23GPT-6 Luna1367
estimated from related benchmarks
~#26/87~1336Estimated
  1. #1Claude Opus 5.51846
  2. #25GPT-5.51336
  3. ~#26Nex-N2.5-MiniEstimated1336
  4. #26MiniMax-M31230
estimated from related benchmarks
~#23/87~1374Estimated
  1. #1Claude Opus 5.51846
  2. #22Qwen3.8 27B1409
  3. ~#23Solar Pro 4Estimated1374
  4. #23GPT-6 Luna1367
estimated from related benchmarks
AutomationBench-AA
third-party
~#29/80~26.8%Estimated
  1. #1Claude Opus 5.569.5%
  2. #28Apodex 1.128.2%
  3. ~#29Ling 3.0 FlashEstimated26.8%
  4. #29GPT-5.4 mini26.2%
estimated from related benchmarks
~#34/80~23.4%Estimated
  1. #1Claude Opus 5.569.5%
  2. #33Quasar 438B23.9%
  3. ~#34Nex-N2.5-MiniEstimated23.4%
  4. #34MiniMax-M321.3%
estimated from related benchmarks
~#31/80~25.6%Estimated
  1. #1Claude Opus 5.569.5%
  2. #30K2 Horizon MoVA 36B A4B26.1%
  3. ~#31Solar Pro 4Estimated25.6%
  4. #31Gemini 3.5 Flash-Lite25.0%
estimated from related benchmarks
τ-Bench Banking (AA)
third-party
#26/8827.2%BestVerified
  1. #1Muse Spark 1.350.5%
  2. #25GPT-5.4 nano27.4%
  3. #26Ling 3.0 Flash27.2%
  4. #27GPT-5.4 mini25.6%
artificialanalysis.ai · 2026-09-24
~#32/89~22.5%Estimated
  1. #1Muse Spark 1.350.5%
  2. #31Hy322.9%
  3. ~#32Nex-N2.5-MiniEstimated22.5%
  4. #32G9v3-39A5B22.1%
estimated from related benchmarks
#30/8823.3%Verified
  1. #1Muse Spark 1.350.5%
  2. #29Muse Glimmer23.5%
  3. #30Solar Pro 423.3%
  4. #31Hy322.9%
artificialanalysis.ai · 2026-09-24
Not comparable across labs (1)
OSWorld 2.0 (release not stated)
day 0
—
#4/430.5%Lab-reported
  1. #1Muse Spark 1.366.9%
  2. #3Qwen3.8 Flash-Next52.3%
  3. #4Nex-N2.5-Mini30.5%
github.com · 2026-09-08
—