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.
MiMo-V2.6-FlashNex-N2.5-ProQwen3.8 27B
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
  • MiMo-V2.6-Flash
    Quality
    63.7
    Rank
    #17/103
    Price
    $0.18/M tok
    Speed
    55tok/s
    In$0.14▾Out$0.28▾/M tok

    vs Nex-N2.5-Pro: +2.1 quality · price n/a

  • Nex-N2.5-Pro
    Quality
    61.6
    Rank
    #21/103
    Price
    —
    Speed
    —

    vs MiMo-V2.6-Flash: −2.1 quality · price n/a

  • Qwen3.8 27B
    Quality
    43.9
    Rank
    #39/103
    Price
    $1.13/M tok
    Speed
    43tok/s
    In$0.50Out$3.00/M tok

    vs MiMo-V2.6-Flash: −19.8 quality · 6.4× price

Overview

Metric
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B
Lab
Xiaomi MiMo
Nex AGI
Alibaba Qwen
Released
Sep 21, 2026
Sep 8, 2026
Aug 14, 2026
Weights
Open
Open
Open
Input price
$/M tok
$0.14Best
—
$0.50
Output price
$/M tok
$0.28Best
—
$3.00
Blended price
$/M tok · 3:1 in:out
$0.175Best
—
$1.125
Output speed
tok/s
55Best
—
43
⚡ Cerebras ~1,850 +1
Context
tokens
1MBest
262K
262K

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
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B
Coding
63.7±6.4Best
#17/103Mixedbest value
61.6±8.6
#21/103Lab-reported
43.9±5.2
#39/103Mixed
Writing & chat
Not ranked
Not ranked
64.3±3.6
#37/146Mixed
Vision
Not ranked
Not ranked
69.0±5.3
#17/82Mixed
Hard reasoning
Not ranked
Not ranked
69.8±3.7
#31/182Verified
Agents
74.8±10.7Best
#19/173Lab-reportedbest value
73.9±8.9
#21/173Lab-reported
74.2±4.7
#20/173Mixed

Coding benchmarks

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

Metric
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B
SWE-bench Verified
day 0
~#3/14~83.5%Estimated
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3MiMo-V2.6-FlashEstimated83.5%
  4. #3Gemini 3.1 Pro80.6%
estimated from related benchmarks
~#3/14~83.2%Estimated
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3Nex-N2.5-ProEstimated83.2%
  4. #3Gemini 3.1 Pro80.6%
estimated from related benchmarks
~#6/14~77.8%Estimated
  1. #1Claude Opus 596.0%
  2. #5Hy378.0%
  3. ~#6Qwen3.8 27BEstimated77.8%
  4. #6Inkling77.6%
estimated from related benchmarks
SWE-bench Pro (Public)
day 0
~#6/19~65.9%Estimated
  1. #1Claude Opus 5.589.9%
  2. #5Intern-S2-397B68.5%
  3. ~#6MiMo-V2.6-FlashEstimated65.9%
  4. #6Hy4 preview65.7%
estimated from related benchmarks
#9/1861.2%Lab-reported
  1. #1Claude Opus 5.589.9%
  2. #8Claude Sonnet 563.2%
  3. #9Nex-N2.5-Pro61.2%
  4. #10Atria Dawn Preview59.6%
github.com · 2026-09-08
~#14/19~57.7%Estimated
  1. #1Claude Opus 5.589.9%
  2. #13Hy357.9%
  3. ~#14Qwen3.8 27BEstimated57.7%
  4. #14Inkling54.3%
estimated from related benchmarks
DeepSWE v1.1
day 0
#10/3067.9%BestLab-reported
  1. #1DeepSeek V4.1 Flash74.2%
  2. #8Kimi K368.4%
  3. #10MiMo-V2.6-Flash67.9%
  4. #11Step 5 Preview67.7%
huggingface.co · 2026-09-21
#26/3055.8%Lab-reported
  1. #1DeepSeek V4.1 Flash74.2%
  2. #25Qwen3.8 2.4T A95B56.6%
  3. #26Nex-N2.5-Pro55.8%
  4. #27Qwen3.8 Max (0902)51.0%
github.com · 2026-09-08
#28/3042.2%Lab-reported
  1. #1DeepSeek V4.1 Flash74.2%
  2. #27Qwen3.8 Max (0902)51.0%
  3. #28Qwen3.8 27B42.2%
  4. #29Nex-N2.5-Mini36.1%
huggingface.co · 2026-08-14
Terminal-Bench 4.0
day 0
#17/8322.7%BestVerified
  1. #1Claude Sonnet 5.563.6%
  2. #16Qwen3.8 Flash-Next25.3%
  3. #17MiMo-V2.6-Flash22.7%
  4. #18Gemini 3.8 Flash19.7%
artificialanalysis.ai · 2026-09-28
~#17/84~24.5%Estimated
  1. #1Claude Sonnet 5.563.6%
  2. #16Qwen3.8 Flash-Next25.3%
  3. ~#17Nex-N2.5-ProEstimated24.5%
  4. #17MiMo-V2.6-Flash22.7%
estimated from related benchmarks
#27/835.6%Verified
  1. #1Claude Sonnet 5.563.6%
  2. #26Qwen3.8 2.4T A95B11.1%
  3. #27Qwen3.8 27B5.6%
  4. #28Gemini 3.1 Pro4.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBench
third-party
~#3/20~91.7%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. ~#3MiMo-V2.6-FlashEstimated91.7%
  4. #3Qwen3.8 27B90.3%
estimated from related benchmarks
~#6/20~87.7%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5gpt-oss-120b87.8%
  3. ~#6Nex-N2.5-ProEstimated87.7%
  4. #6Gemma 4 31B80.0%
estimated from related benchmarks
#3/1990.3%Lab-reported
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #3Qwen3.8 27B90.3%
  4. #4Nemotron 3 Ultra89.0%
huggingface.co · 2026-08-14
SciCode
third-party
#26/9551.3%BestVerified
  1. #1Claude Opus 5.566.9%
  2. #25GLM-5.3 Flash51.6%
  3. #26MiMo-V2.6-Flash51.3%
  4. #27DeepSeek V4 Pro (0813)51.0%
artificialanalysis.ai · 2026-09-28
~#11/96~57.5%Estimated
  1. #1Claude Opus 5.566.9%
  2. #10GPT-6 Sol57.6%
  3. ~#11Nex-N2.5-ProEstimated57.5%
  4. #11Grok 4.757.4%
estimated from related benchmarks
#38/9546.6%Verified
  1. #1Claude Opus 5.566.9%
  2. #37Inkling47.0%
  3. #38Qwen3.8 27B46.6%
  4. #39Qwen3.7 Plus46.1%
artificialanalysis.ai · 2026-09-24
AA Coding Agent Index
third-party
~#10/15~48.2Estimated
  1. #1Claude Opus 5.566.0
  2. #9Kimi K351.9
  3. ~#10MiMo-V2.6-FlashEstimated48.2
  4. #10Grok 4.647.0
estimated from related benchmarks
~#10/15~51.6Estimated
  1. #1Claude Opus 5.566.0
  2. #9Kimi K351.9
  3. ~#10Nex-N2.5-ProEstimated51.6
  4. #10Grok 4.647.0
estimated from related benchmarks
~#14/15~41.2Estimated
  1. #1Claude Opus 5.566.0
  2. #13Gemini 3.8 Flash41.9
  3. ~#14Qwen3.8 27BEstimated41.2
  4. #14GPT-6 Luna41.1
estimated from related benchmarks
Not comparable across labs (1)
SWE-bench Pro (Qwen-corrected)
day 0
—
—
#4/461.7%Lab-reported
  1. #1Qwen3.8 2.4T A95B67.7%
  2. #3Qwen3.8 Flash-Next62.5%
  3. #4Qwen3.8 27B61.7%
huggingface.co · 2026-08-14
Benchmarks in other categories (4)

Writing & chat benchmarks

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

Metric
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B

Vision benchmarks

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

Metric
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B
MMMU
day 0
—
—
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2Qwen3.8 27BEstimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
MMMU-Pro
day 0
—
—
#26/7676.3%Verified
  1. #1Claude Opus 5.587.7%
  2. #25Step 5 Preview76.4%
  3. #26Qwen3.8 27B76.3%
  4. #27GPT-6 Luna75.5%
artificialanalysis.ai · 2026-09-24
CharXiv Reasoning
day 0
—
—
~#5/9~80.4%Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #4Qwen3.8-Omni-Flash83.5%
  3. ~#5Qwen3.8 27BEstimated80.4%
  4. #5Muse Glimmer78.8%
estimated from related benchmarks
OmniDocBench v1.5
third-party
—
—
#2/791.1%Lab-reported
  1. #1MiniMax-M391.6%
  2. #2Kimi K391.1%
  3. #2Qwen3.8 27B91.1%
  4. #4Gemini 3.1 Pro85.3%
huggingface.co · 2026-08-14
GDP.pdf (AA)
third-party
—
—
#17/4216.6%Verified
  1. #1GPT-6 Astra31.0%
  2. #16Grok 4.617.2%
  3. #17Qwen3.8 27B16.6%
  4. #18Qwen3.8 Flash-Next15.6%
artificialanalysis.ai · 2026-09-24
Chartography
third-party
—
—
~#14/23~24.1%Estimated
  1. #1GPT-6 Astra71.0%
  2. #13Gemini 3.1 Pro26.1%
  3. ~#14Qwen3.8 27BEstimated24.1%
  4. #14Grok 4.620.5%
estimated from related benchmarks
Not comparable across labs (1)
CharXiv Reasoning (lab-corrected key)
day 0
—
—
#1/183.7%Lab-reported
huggingface.co · 2026-08-14

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
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B

Agents benchmarks

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

Metric
MiMo-V2.6-Flash
Nex-N2.5-Pro
Qwen3.8 27B
τ²-bench (Telecom)
day 0
~#2/116~96.0%Estimated
  1. #1Step 3.7 Flash98.5%
  2. ~#2MiMo-V2.6-FlashEstimated96.0%
  3. #2Gemini 3.1 Pro95.6%
estimated from related benchmarks
~#8/116~93.1%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #7Qwen3.5 122B A10B93.6%
  3. ~#8Nex-N2.5-ProEstimated93.1%
  4. #8Qwen3.7 Plus93.0%
estimated from related benchmarks
~#8/116~93.3%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #7Qwen3.5 122B A10B93.6%
  3. ~#8Qwen3.8 27BEstimated93.3%
  4. #8Qwen3.7 Plus93.0%
estimated from related benchmarks
OSWorld-Verified
day 0
#6/1180.8%Lab-reported
  1. #1Kimi K384.8%
  2. #5Claude Sonnet 581.2%
  3. #6MiMo-V2.6-Flash80.8%
  4. #7GPT-5.578.7%
huggingface.co · 2026-09-21
#3/1182.2%Lab-reported
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #3Nex-N2.5-Pro82.2%
  4. #4MiMo-V2.6-Pro82.0%
github.com · 2026-09-08
#2/1184.3%BestLab-reported
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #3Nex-N2.5-Pro82.2%
huggingface.co · 2026-08-14
OSWorld 2.0
day 0
~#3/5~53.8%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3MiMo-V2.6-FlashEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
~#3/5~55.3%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Nex-N2.5-ProEstimated55.3%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
~#3/5~53.6%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Qwen3.8 27BEstimated53.6%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
BrowseComp
day 0
~#8/18~86.4%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #7GPT-5.6 Terra87.5%
  3. ~#8MiMo-V2.6-FlashEstimated86.4%
  4. #8Gemini 3.1 Pro85.9%
estimated from related benchmarks
#5/1789.7%Lab-reported
  1. #1Atria Dawn Preview92.5%
  2. #4Claude Opus 590.8%
  3. #5Nex-N2.5-Pro89.7%
  4. #6Step 5 Preview88.7%
github.com · 2026-09-08
~#9/18~84.9%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #8Gemini 3.1 Pro85.9%
  3. ~#9Qwen3.8 27BEstimated84.9%
  4. #9Claude Sonnet 584.7%
estimated from related benchmarks
GDPval-AA (Elo)
third-party
~#16/87~1538Estimated
  1. #1Claude Opus 5.51846
  2. #15GPT-6 Astra1542
  3. ~#16MiMo-V2.6-FlashEstimated1538
  4. #16Kimi K31524
estimated from related benchmarks
~#17/87~1492Estimated
  1. #1Claude Opus 5.51846
  2. #16Kimi K31524
  3. ~#17Nex-N2.5-ProEstimated1492
  4. #17GPT-6 Sol1487
estimated from related benchmarks
#22/861409Verified
  1. #1Claude Opus 5.51846
  2. #21Gemini 3.8 Flash1412
  3. #22Qwen3.8 27B1409
  4. #23GPT-6 Luna1367
artificialanalysis.ai · 2026-09-24
AutomationBench-AA
third-party
~#22/80~49.0%Estimated
  1. #1Claude Opus 5.569.5%
  2. #21Step 5 Preview51.0%
  3. ~#22MiMo-V2.6-FlashEstimated49.0%
  4. #22Qwen3.8 27B48.2%
estimated from related benchmarks
~#24/80~43.5%Estimated
  1. #1Claude Opus 5.569.5%
  2. #23GPT-5.547.3%
  3. ~#24Nex-N2.5-ProEstimated43.5%
  4. #24K2 Horizon 375B A23B37.2%
estimated from related benchmarks
#22/7948.2%Verified
  1. #1Claude Opus 5.569.5%
  2. #21Step 5 Preview51.0%
  3. #22Qwen3.8 27B48.2%
  4. #23GPT-5.547.3%
artificialanalysis.ai · 2026-09-24
τ-Bench Banking (AA)
third-party
~#17/89~37.4%Estimated
  1. #1Muse Spark 1.350.5%
  2. #16GPT-5.539.0%
  3. ~#17MiMo-V2.6-FlashEstimated37.4%
  4. #17Claude Sonnet 537.3%
estimated from related benchmarks
~#19/89~34.7%Estimated
  1. #1Muse Spark 1.350.5%
  2. #18Motif 335.3%
  3. ~#19Nex-N2.5-ProEstimated34.7%
  4. #19Ling-3.0-flash-VL34.4%
estimated from related benchmarks
#4/8848.0%Verified
  1. #1Muse Spark 1.350.5%
  2. #3Qwen3.8 2.4T A95B49.1%
  3. #4Qwen3.8 27B48.0%
  4. #5Qwen3.8 Max (0902)47.8%
artificialanalysis.ai · 2026-09-24
Not comparable across labs (1)
OSWorld 2.0 (release not stated)
day 0
—
#2/456.4%Lab-reported
  1. #1Muse Spark 1.366.9%
  2. #2Nex-N2.5-Pro56.4%
  3. #3Qwen3.8 Flash-Next52.3%
github.com · 2026-09-08
—