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.
Motif 3Claude Sonnet 5GPT-6 Luna
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
  • Motif 3
    Quality
    41.2
    Rank
    #40/103
    Price
    —
    Speed
    —

    vs Claude Sonnet 5: −15.4 quality · price n/a

  • Claude Sonnet 5
    Quality
    56.6
    Rank
    #27/103
    Price
    $4.00/M tok
    Speed
    79tok/s
    In$2.00Out$10/M tok

    vs GPT-6 Luna: +4.2 quality · 20× price

  • GPT-6 Luna
    Quality
    52.5
    Rank
    #30/103
    Price
    $0.20/M tok
    Speed
    132tok/s
    In$0.10▾Out$0.50▾/M tok

    vs Claude Sonnet 5: −4.2 quality · 0.05× price

Overview

Metric
Motif 3
Claude Sonnet 5
GPT-6 Luna
Lab
Motif Technologies
Anthropic
OpenAI
Released
Aug 12, 2026
Jun 30, 2026
Sep 22, 2026
Weights
Open
Proprietary
Proprietary
Input price
$/M tok
—
$2.00
$0.10Best
Output price
$/M tok
—
$10
$0.50Best
Blended price
$/M tok · 3:1 in:out
—
$4.00
$0.20Best
Output speed
tok/s
—
79
132Best
Context
tokens
262K
1MBest
1MBest

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
Motif 3
Claude Sonnet 5
GPT-6 Luna
Coding
41.2±10.0
#40/103Verified
56.6±5.4Best
#27/103Mixed
52.5±4.5
#30/103Verified
Writing & chat
Not ranked
68.3±4.3Best
#29/146Verified
67.2±6.0
#33/146Verifiedbest value
Vision
Not ranked
56.2±4.2
#36/82Mixed
68.8±4.8Best
#18/82Verifiedbest value
Hard reasoning
69.3±5.8
#32/182Verified
76.5±3.7Best
#22/182Verified
72.4±5.1
#28/182Verifiedbest value
Agents
65.2±10.0
#28/173Verified
65.3±3.8Best
#27/173Mixed
64.2±7.8
#29/173Verified

Coding benchmarks

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

Metric
Motif 3
Claude Sonnet 5
GPT-6 Luna
SWE-bench Verified
day 0
~#10/14~74.5%Estimated
  1. #1Claude Opus 596.0%
  2. #9Xing4.0-29B-A4B75.0%
  3. ~#10Motif 3Estimated74.5%
  4. #10Claude Haiku 4.573.3%
estimated from related benchmarks
#2/1385.2%Lab-reported
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. #3Gemini 3.1 Pro80.6%
www-cdn.anthropic.com · 2026-06-30
~#3/14~82.0%Estimated
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3GPT-6 LunaEstimated82.0%
  4. #3Gemini 3.1 Pro80.6%
estimated from related benchmarks
SWE-bench Pro (Public)
day 0
~#17/19~54.0%Estimated
  1. #1Claude Opus 5.589.9%
  2. #15Gemini 3.5 Flash-Lite54.2%
  3. ~#17Motif 3Estimated54.0%
  4. #17Muse Glimmer51.2%
estimated from related benchmarks
#8/1863.2%Lab-reported
  1. #1Claude Opus 5.589.9%
  2. #7GPT-5.6 Terra63.4%
  3. #8Claude Sonnet 563.2%
  4. #9Nex-N2.5-Pro61.2%
anthropic.com · 2026-06-30
~#9/19~62.5%Estimated
  1. #1Claude Opus 5.589.9%
  2. #8Claude Sonnet 563.2%
  3. ~#9GPT-6 LunaEstimated62.5%
  4. #9Nex-N2.5-Pro61.2%
estimated from related benchmarks
DeepSWE v1.1
day 0
~#28/31~47.7%Estimated
  1. #1DeepSeek V4.1 Flash74.2%
  2. #27Qwen3.8 Max (0902)51.0%
  3. ~#28Motif 3Estimated47.7%
  4. #28Qwen3.8 27B42.2%
estimated from related benchmarks
~#22/31~59.1%Estimated
  1. #1DeepSeek V4.1 Flash74.2%
  2. #21GLM-5.361.4%
  3. ~#22Claude Sonnet 5Estimated59.1%
  4. #22Qwen3.8 Flash-Next58.7%
estimated from related benchmarks
#18/3063.7%Verified
  1. #1DeepSeek V4.1 Flash74.2%
  2. #16Hy4 preview64.3%
  3. #18GPT-6 Luna63.7%
  4. #19GLM-5.3 Flash63.4%
artificialanalysis.ai · 2026-09-24
Terminal-Bench 4.0
day 0
~#23/84~12.9%Estimated
  1. #1Claude Sonnet 5.563.6%
  2. #21DeepSeek V4 Pro (0813)14.1%
  3. ~#23Motif 3Estimated12.9%
  4. #23GPT-5.5 Instant (June 2026)12.6%
estimated from related benchmarks
#21/8314.1%BestVerified
  1. #1Claude Sonnet 5.563.6%
  2. #20GPT-5.514.6%
  3. #21Claude Sonnet 514.1%
  4. #21DeepSeek V4 Pro (0813)14.1%
artificialanalysis.ai · 2026-09-24
#23/8312.6%Verified
  1. #1Claude Sonnet 5.563.6%
  2. #23GPT-5.5 Instant (June 2026)12.6%
  3. #23GPT-6 Luna12.6%
  4. #23Kimi K312.6%
artificialanalysis.ai · 2026-09-24
LiveCodeBench
third-party
~#7/20~77.9%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #6Gemma 4 31B80.0%
  3. ~#7Motif 3Estimated77.9%
  4. #7gpt-oss-20b77.7%
estimated from related benchmarks
~#6/20~85.9%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5gpt-oss-120b87.8%
  3. ~#6Claude Sonnet 5Estimated85.9%
  4. #6Gemma 4 31B80.0%
estimated from related benchmarks
~#6/20~85.4%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5gpt-oss-120b87.8%
  3. ~#6GPT-6 LunaEstimated85.4%
  4. #6Gemma 4 31B80.0%
estimated from related benchmarks
SciCode
third-party
#52/9542.2%Verified
  1. #1Claude Opus 5.566.9%
  2. #52Claude Haiku 4.542.2%
  3. #52Motif 342.2%
  4. #54K-EXAONE 2.0 080342.0%
artificialanalysis.ai · 2026-09-24
#18/9554.3%Verified
  1. #1Claude Opus 5.566.9%
  2. #17GPT-6 Luna54.6%
  3. #18Claude Sonnet 554.3%
  4. #19Qwen3.8 2.4T A95B54.1%
artificialanalysis.ai · 2026-09-24
#17/9554.6%BestVerified
  1. #1Claude Opus 5.566.9%
  2. #16GPT-5.6 Terra55.0%
  3. #17GPT-6 Luna54.6%
  4. #18Claude Sonnet 554.3%
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. ~#15Motif 3Estimated<41.1
estimated from related benchmarks
~#11/15~45.7Estimated
  1. #1Claude Opus 5.566.0
  2. #10Grok 4.647.0
  3. ~#11Claude Sonnet 5Estimated45.7
  4. #11Qwen3.8 Max (0902)43.3
estimated from related benchmarks
#14/1441.1Verified
  1. #1Claude Opus 5.566.0
  2. #13Gemini 3.8 Flash41.9
  3. #14GPT-6 Luna41.1
artificialanalysis.ai · 2026-09-24
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
Motif 3
Claude Sonnet 5
GPT-6 Luna
IFBench
day 0
—
~#21/126~73.5%Estimated
  1. #1MiniMax-M382.9%
  2. #20Command A+73.9%
  3. ~#21Claude Sonnet 5Estimated73.5%
  4. #21GPT-5.4 mini73.3%
estimated from related benchmarks
~#23/126~71.8%Estimated
  1. #1MiniMax-M382.9%
  2. #22Gemma 4 26B A4B72.4%
  3. ~#23GPT-6 LunaEstimated71.8%
  4. #23Nemotron 3 Super 120B A12B71.5%
estimated from related benchmarks
MMMLU
day 0
—
~#2/5~89.9%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. ~#2Claude Sonnet 5Estimated89.9%
  3. #2Gemma 4 31B88.4%
estimated from related benchmarks
~#2/5~88.4%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. ~#2GPT-6 LunaEstimated88.4%
  3. #2Gemma 4 31B88.4%
estimated from related benchmarks
LMArena Text, Non-English (Elo)
third-party
—
#18/531447BestVerified
  1. #1Claude Opus 5.51501
  2. #16GPT-6 Sol1448
  3. #18Claude Sonnet 51447
  4. #18Grok 4.61447
lmarena.ai · 2026-09-24
#25/531436Verified
  1. #1Claude Opus 5.51501
  2. #24Gemma 4 31BFew votes1437
  3. #25GPT-6 Luna1436
  4. #26Grok 4.71435
lmarena.ai · 2026-09-25
LMArena Text (Elo)
third-party
—
#17/531461BestVerified
  1. #1Claude Opus 5.51509
  2. #16DeepSeek V4 Pro (0813)1463
  3. #17Claude Sonnet 51461
  4. #18GPT-6 Sol1457
lmarena.ai · 2026-09-24
#25/531442Verified
  1. #1Claude Opus 5.51509
  2. #24GPT-5.4 mini1448
  3. #25GPT-6 Luna1442
  4. #25Qwen3.5 397B A17B1442
lmarena.ai · 2026-09-25
EQ-Bench Creative Writing v3 (Elo)
third-party
—
#14/251794Verified
  1. #1GPT-6 Astra2173
  2. #13Muse Glimmer1798
  3. #14Claude Sonnet 51794
  4. #15Gemini 3.8 Flash1748
eqbench.com · 2026-09-24
~#10/26~1862Estimated
  1. #1GPT-6 Astra2173
  2. #9Muse Spark 1.31906
  3. ~#10GPT-6 LunaEstimated1862
  4. #10GPT-5.6 Terra1855
estimated from related benchmarks

Vision benchmarks

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

Metric
Motif 3
Claude Sonnet 5
GPT-6 Luna
MMMU
day 0
—
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2Claude Sonnet 5Estimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2GPT-6 LunaEstimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
MMMU-Pro
day 0
—
#20/7677.3%BestVerified
  1. #1Claude Opus 5.587.7%
  2. #19GPT-5.3 Codex78.5%
  3. #20Claude Sonnet 577.3%
  4. #20Qwen3.5 397B A17B77.3%
artificialanalysis.ai · 2026-09-24
#27/7675.5%Verified
  1. #1Claude Opus 5.587.7%
  2. #26Qwen3.8 27B76.3%
  3. #27GPT-6 Luna75.5%
  4. #28MiMo-V2.575.4%
artificialanalysis.ai · 2026-09-24
CharXiv Reasoning
day 0
—
#7/877.0%Lab-reported
  1. #1Gemini 3.8 Flash86.2%
  2. #6Inkling78.1%
  3. #7Claude Sonnet 577.0%
  4. #8Gemini 3.5 Flash-Lite74.5%
www-cdn.anthropic.com · 2026-06-30
~#5/9~81.5%Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #4Qwen3.8-Omni-Flash83.5%
  3. ~#5GPT-6 LunaEstimated81.5%
  4. #5Muse Glimmer78.8%
estimated from related benchmarks
OmniDocBench v1.5
third-party
—
~#4/8~85.7%Estimated
  1. #1MiniMax-M391.6%
  2. #2Qwen3.8 27B91.1%
  3. ~#4Claude Sonnet 5Estimated85.7%
  4. #4Gemini 3.1 Pro85.3%
estimated from related benchmarks
~#4/8~87.3%Estimated
  1. #1MiniMax-M391.6%
  2. #2Qwen3.8 27B91.1%
  3. ~#4GPT-6 LunaEstimated87.3%
  4. #4Gemini 3.1 Pro85.3%
estimated from related benchmarks
GDP.pdf (AA)
third-party
—
#23/4213.2%Verified
  1. #1GPT-6 Astra31.0%
  2. #22GPT-5.4 mini13.4%
  3. #23Claude Sonnet 513.2%
  4. #24DeepSeek V4.1 Flash12.8%
artificialanalysis.ai · 2026-09-24
#12/4220.4%BestVerified
  1. #1GPT-6 Astra31.0%
  2. #11Gemini 3.8 Flash21.0%
  3. #12GPT-6 Luna20.4%
  4. #13Grok 4.720.0%
artificialanalysis.ai · 2026-09-24
Chartography
third-party
—
#16/2216.6%Verified
  1. #1GPT-6 Astra71.0%
  2. #15Muse Glimmer17.2%
  3. #16Claude Sonnet 516.6%
  4. #17GLM-5.3 Flash16.3%
surgehq.ai · 2026-09-24
#8/2229.1%BestVerified
  1. #1GPT-6 Astra71.0%
  2. #7GPT-5.531.0%
  3. #8GPT-6 Luna29.1%
  4. #8Qwen3.8 Max (0902)29.1%
surgehq.ai · 2026-09-24

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
Motif 3
Claude Sonnet 5
GPT-6 Luna
GPQA Diamond
day 0
#51/17283.4%Verified
  1. #1GPT-6 Astra96.1%
  2. #50Muse Glimmer83.5%
  3. #51Motif 383.4%
  4. #52K-EXAONE 2.0 080382.9%
artificialanalysis.ai · 2026-09-24
#21/17291.1%BestVerified
  1. #1GPT-6 Astra96.1%
  2. #20GLM-5.3 Flash91.2%
  3. #21Claude Sonnet 591.1%
  4. #22Qwen3.8-Omni-Flash91.0%
artificialanalysis.ai · 2026-09-24
~#33/173~87.9%Estimated
  1. #1GPT-6 Astra96.1%
  2. #32Solar Pro 489.1%
  3. ~#33GPT-6 LunaEstimated87.9%
  4. #33GPT-5.4 mini87.5%
estimated from related benchmarks
Humanity's Last Exam
day 0
#30/17837.0%Verified
  1. #1Claude Opus 5.561.4%
  2. #29Qwen3.8 Flash-Next38.0%
  3. #30Motif 337.0%
  4. #31Qwen3.8-Omni-Flash36.5%
artificialanalysis.ai · 2026-09-24
#22/17841.3%BestVerified
  1. #1Claude Opus 5.561.4%
  2. #21GLM-5.342.3%
  3. #22Claude Sonnet 541.3%
  4. #23DeepSeek V4 Pro (0813)41.0%
artificialanalysis.ai · 2026-09-24
#27/17838.5%Verified
  1. #1Claude Opus 5.561.4%
  2. #26MiniMax-M339.0%
  3. #27GPT-6 Luna38.5%
  4. #28Grok Build 0.1 061638.3%
artificialanalysis.ai · 2026-09-24
AIME (latest)
day 0
~#4/25~94.2%Estimated
  1. #1Inkling97.1%
  2. #3Nova 2.0 Lite94.3%
  3. ~#4Motif 3Estimated94.2%
  4. #4gpt-oss-120b93.4%
estimated from related benchmarks
~#2/25~95.9%Estimated
  1. #1Inkling97.1%
  2. ~#2Claude Sonnet 5Estimated95.9%
  3. #2Muse Glimmer94.7%
estimated from related benchmarks
~#3/25~94.4%Estimated
  1. #1Inkling97.1%
  2. #2Muse Glimmer94.7%
  3. ~#3GPT-6 LunaEstimated94.4%
  4. #3Nova 2.0 Lite94.3%
estimated from related benchmarks
AA Intelligence Index
third-party
~#26/75~32.0Estimated
  1. #1Claude Opus 5.557.6
  2. #25Qwen3.8 27B33.7
  3. ~#26Motif 3Estimated32.0
  4. #26K2 Horizon 375B A23B30.5
estimated from related benchmarks
#22/7438.2BestVerified
  1. #1Claude Opus 5.557.6
  2. #21GPT-5.538.4
  3. #22Claude Sonnet 538.2
  4. #23GPT-6 Luna37.3
artificialanalysis.ai · 2026-09-24
#23/7437.3Verified
  1. #1Claude Opus 5.557.6
  2. #22Claude Sonnet 538.2
  3. #23GPT-6 Luna37.3
  4. #24DeepSeek V4 Pro (0813)36.0
artificialanalysis.ai · 2026-09-24

Agents benchmarks

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

Metric
Motif 3
Claude Sonnet 5
GPT-6 Luna
τ²-bench (Telecom)
day 0
~#2/116~96.0%Estimated
  1. #1Step 3.7 Flash98.5%
  2. ~#2Motif 3Estimated96.0%
  3. #2Gemini 3.1 Pro95.6%
estimated from related benchmarks
~#11/116~91.6%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #10Ring-2.6-1T92.4%
  3. ~#11Claude Sonnet 5Estimated91.6%
  4. #11MiMo-V2.590.6%
estimated from related benchmarks
~#16/116~89.2%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #15KAT Coder Pro V289.5%
  3. ~#16GPT-6 LunaEstimated89.2%
  4. #16MiniMax-M388.9%
estimated from related benchmarks
OSWorld-Verified
day 0
~#8/12~77.3%Estimated
  1. #1Kimi K384.8%
  2. #7GPT-5.578.7%
  3. ~#8Motif 3Estimated77.3%
  4. #8Gemini 3.5 Flash-Lite74.0%
estimated from related benchmarks
#5/1181.2%Lab-reported
  1. #1Kimi K384.8%
  2. #4MiMo-V2.6-Pro82.0%
  3. #5Claude Sonnet 581.2%
  4. #6MiMo-V2.6-Flash80.8%
anthropic.com · 2026-06-30
~#8/12~78.7%Estimated
  1. #1Kimi K384.8%
  2. #7GPT-5.578.7%
  3. ~#8GPT-6 LunaEstimated78.7%
  4. #8Gemini 3.5 Flash-Lite74.0%
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. ~#5Motif 3Estimated<47.5%
estimated from related benchmarks
~last/5below measured range (<47.5%)Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #4GPT-5.547.5%
  3. ~#5Claude Sonnet 5Estimated<47.5%
estimated from related benchmarks
~#3/5~50.4%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3GPT-6 LunaEstimated50.4%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
BrowseComp
day 0
~#14/18~82.9%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #13Nex-N2.5-Mini83.4%
  3. ~#14Motif 3Estimated82.9%
  4. #14Inkling77.1%
estimated from related benchmarks
#9/1784.7%Lab-reported
  1. #1Atria Dawn Preview92.5%
  2. #8Gemini 3.1 Pro85.9%
  3. #9Claude Sonnet 584.7%
  4. #10GPT-5.584.4%
www-cdn.anthropic.com · 2026-06-30
~#14/18~81.8%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #13Nex-N2.5-Mini83.4%
  3. ~#14GPT-6 LunaEstimated81.8%
  4. #14Inkling77.1%
estimated from related benchmarks
GDPval-AA (Elo)
third-party
~#18/87~1484Estimated
  1. #1Claude Opus 5.51846
  2. #17GPT-6 Sol1487
  3. ~#18Motif 3Estimated1484
  4. #18Claude Sonnet 51449
estimated from related benchmarks
#18/861449BestVerified
  1. #1Claude Opus 5.51846
  2. #17GPT-6 Sol1487
  3. #18Claude Sonnet 51449
  4. #19DeepSeek V4 Pro (0813)1441
artificialanalysis.ai · 2026-09-24
#23/861367Verified
  1. #1Claude Opus 5.51846
  2. #22Qwen3.8 27B1409
  3. #23GPT-6 Luna1367
  4. #24K2 Horizon 375B A23B1349
artificialanalysis.ai · 2026-09-24
AutomationBench-AA
third-party
~#24/80~39.7%Estimated
  1. #1Claude Opus 5.569.5%
  2. #23GPT-5.547.3%
  3. ~#24Motif 3Estimated39.7%
  4. #24K2 Horizon 375B A23B37.2%
estimated from related benchmarks
#25/7936.5%Verified
  1. #1Claude Opus 5.569.5%
  2. #24K2 Horizon 375B A23B37.2%
  3. #25Claude Sonnet 536.5%
  4. #26GPT-5.5 Instant (June 2026)36.4%
artificialanalysis.ai · 2026-09-24
#20/7953.2%BestVerified
  1. #1Claude Opus 5.569.5%
  2. #19Qwen3.8 Flash-Next55.9%
  3. #20GPT-6 Luna53.2%
  4. #21Step 5 Preview51.0%
artificialanalysis.ai · 2026-09-24
τ-Bench Banking (AA)
third-party
#18/8835.3%Verified
  1. #1Muse Spark 1.350.5%
  2. #17Claude Sonnet 537.3%
  3. #18Motif 335.3%
  4. #19Ling-3.0-flash-VL34.4%
artificialanalysis.ai · 2026-09-24
#17/8837.3%BestVerified
  1. #1Muse Spark 1.350.5%
  2. #16GPT-5.539.0%
  3. #17Claude Sonnet 537.3%
  4. #18Motif 335.3%
artificialanalysis.ai · 2026-09-24
~#22/89~31.5%Estimated
  1. #1Muse Spark 1.350.5%
  2. #21K2 Horizon MoVA 36B A4B32.0%
  3. ~#22GPT-6 LunaEstimated31.5%
  4. #22Inkling29.1%
estimated from related benchmarks
Not comparable across labs (1)
OSWorld 2.0 (offline set v2026.08.08)
day 0
—
—
#3/352.7%Lab-reported
  1. #1GPT-6 Astra72.6%
  2. #2GPT-6 Sol64.4%
  3. #3GPT-6 Luna52.7%
openai.com · 2026-09-22