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.
GLM-5.3 Flash NVFP4Grok 4.7Ternary Bonsai 2 27B
Focus

Coding: quality vs price

Compared models are ringed; the other 111 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, Qwen3.8-Omni-Flash, GLM-5.3 Flash, DeepSeek V4.1 Flash, MiMo-V2.6-Pro, Claude Sonnet 5.5, Claude Opus 5.5. 36 models are not plotted: GLM-5.3 Flash NVFP4, Ternary Bonsai 2 27B, DeepSeek V4.1 Flash NVFP4, GLM-5.3 NVFP4, Nex-N2.5-Pro, Xing4.0-29B-A4B, Atria Dawn Preview, Ternary Bonsai 27B, Maple-Preview, Bonsai 27B, Qwen3.8 27B NVFP4, 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, Qwen3.8 Flash-Next NVFP4, Devstral 2, K2 Horizon 7B, K2-Horizon-32B, Devstral Small 2, MiniCPM5-2B, Ling 3.0 Tiny, K2 Horizon 3.7B, K2 Horizon 0.9B, LFM2.5-2.6B.

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

GLM-5.3 Flash NVFP4, Ternary Bonsai 2 27B, DeepSeek V4.1 Flash NVFP4, GLM-5.3 NVFP4, Nex-N2.5-Pro, Xing4.0-29B-A4B, Atria Dawn Preview, Ternary Bonsai 27B, Maple-Preview, Bonsai 27B, Qwen3.8 27B NVFP4, 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, Qwen3.8 Flash-Next NVFP4, Devstral 2, K2 Horizon 7B, K2-Horizon-32B, 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
  • GLM-5.3 Flash NVFP4
    Quality
    71.8
    Rank
    #12/114
    Price
    —
    Speed
    —

    vs Ternary Bonsai 2 27B: ±0.0 quality · price n/a

  • Grok 4.7
    Quality
    71.6
    Rank
    #14/114
    Price
    $3.00/M tok
    Speed
    76tok/s
    In$2.00Out$6.00/M tok

    vs GLM-5.3 Flash NVFP4: −0.2 quality · price n/a

  • Ternary Bonsai 2 27B
    Quality
    71.8
    Rank
    #13/114
    Price
    —
    Speed
    —

    vs GLM-5.3 Flash NVFP4: ±0.0 quality · price n/a

Overview

Metric
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
Lab
NVIDIA
SpaceXAI (xAI)
PrismML
Released
Sep 2, 2026
Sep 21, 2026
Sep 16, 2026
Weights
Open
Proprietary
Open
Input price
$/M tok
—
$2.00
—
Output price
$/M tok
—
$6.00
—
Blended price
$/M tok · 3:1 in:out
—
$3.00
—
Output speed
tok/s
—
76
—
Context
tokens
1MBest
500K
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
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
Coding
71.8±10.1Best
#12/114Lab-reported
71.6±4.5
#14/114Verified
71.8±10.3
#13/114Lab-reported
Writing & chat
67.4±9.7
#37/155Lab-reported
70.2±4.3
#31/155Verified
76.9±9.7Best
#21/155Lab-reported
Vision
68.8±8.7
#22/90Lab-reported
63.4±7.3
#31/90Verified
69.0±8.7Best
#20/90Lab-reported
Hard reasoning
73.0±9.6
#31/193Lab-reported
81.2±5.1Best
#15/193Verified
74.4±9.9
#28/193Lab-reported
Agents
Not ranked
86.5±7.8
#5/176Verified
Not ranked

Coding benchmarks

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

Metric
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
SWE-bench Verified
day 0
~#2/14~87.9%Estimated
  1. #1Claude Opus 596.0%
  2. ~#2GLM-5.3 Flash NVFP4Estimated87.9%
  3. #2Claude Sonnet 585.2%
estimated from related benchmarks
~#2/14~85.7%Estimated
  1. #1Claude Opus 596.0%
  2. ~#2Grok 4.7Estimated85.7%
  3. #2Claude Sonnet 585.2%
estimated from related benchmarks
~#3/14~82.8%Estimated
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3Ternary Bonsai 2 27BEstimated82.8%
  4. #3Gemini 3.1 Pro80.6%
estimated from related benchmarks
SWE-bench Pro (Public)
day 0
~#5/19~70.4%Estimated
  1. #1Claude Opus 5.589.9%
  2. #4Claude Opus 579.2%
  3. ~#5GLM-5.3 Flash NVFP4Estimated70.4%
  4. #5Intern-S2-397B68.5%
estimated from related benchmarks
~#6/19~68.1%Estimated
  1. #1Claude Opus 5.589.9%
  2. #5Intern-S2-397B68.5%
  3. ~#6Grok 4.7Estimated68.1%
  4. #6Hy4 preview65.7%
estimated from related benchmarks
~#5/19~68.7%Estimated
  1. #1Claude Opus 5.589.9%
  2. #4Claude Opus 579.2%
  3. ~#5Ternary Bonsai 2 27BEstimated68.7%
  4. #5Intern-S2-397B68.5%
estimated from related benchmarks
DeepSWE v1.1
day 0
~#15/32~66.1%Estimated
  1. #1GPT-6.1 Sol75.2%
  2. #14GPT-5.567.0%
  3. ~#15GLM-5.3 Flash NVFP4Estimated66.1%
  4. #15Gemini 3.8 Flash65.8%
estimated from related benchmarks
#3/3172.6%Verified
  1. #1GPT-6.1 Sol75.2%
  2. #2DeepSeek V4.1 Flash74.2%
  3. #3Grok 4.772.6%
  4. #4MiMo-V2.6-Pro71.9%
artificialanalysis.ai · 2026-09-24
~#25/32~57.4%Estimated
  1. #1GPT-6.1 Sol75.2%
  2. #24Qwen3.8-Omni-Flash57.8%
  3. ~#25Ternary Bonsai 2 27BEstimated57.4%
  4. #25DeepSeek V4 Pro (0813)57.2%
estimated from related benchmarks
Terminal-Bench 4.0
day 0
~#16/85~26.2%Estimated
  1. #1Claude Sonnet 5.563.6%
  2. #15DeepSeek V4.1 Flash26.8%
  3. ~#16GLM-5.3 Flash NVFP4Estimated26.2%
  4. #16Grok 4.725.8%
estimated from related benchmarks
#16/8425.8%Verified
  1. #1Claude Sonnet 5.563.6%
  2. #15DeepSeek V4.1 Flash26.8%
  3. #16Grok 4.725.8%
  4. #17Qwen3.8 Flash-Next25.3%
artificialanalysis.ai · 2026-09-24
~#15/85~29.6%Estimated
  1. #1Claude Sonnet 5.563.6%
  2. #14GLM-5.3 Flash32.8%
  3. ~#15Ternary Bonsai 2 27BEstimated29.6%
  4. #15DeepSeek V4.1 Flash26.8%
estimated from related benchmarks
LiveCodeBench
third-party
~#6/25~88.7%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5Nemotron 3 Ultra89.0%
  3. ~#6GLM-5.3 Flash NVFP4Estimated88.7%
  4. #6gpt-oss-120b87.8%
estimated from related benchmarks
~#7/25~83.3%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #6gpt-oss-120b87.8%
  3. ~#7Grok 4.7Estimated83.3%
  4. #7Ternary Bonsai 27B82.8%
estimated from related benchmarks
#4/2490.1%Lab-reported
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #3Qwen3.8 27B90.3%
  3. #4Ternary Bonsai 2 27B90.1%
  4. #5Nemotron 3 Ultra89.0%
huggingface.co · 2026-09-29
SciCode
third-party
#11/10257.7%BestLab-reported
  1. #1Claude Opus 5.566.9%
  2. #10GLM-5.3 NVFP458.0%
  3. #11GLM-5.3 Flash NVFP457.7%
  4. #12GPT-6 Sol57.6%
huggingface.co · 2026-09-29
#13/10257.4%Verified
  1. #1Claude Opus 5.566.9%
  2. #12GPT-6 Sol57.6%
  3. #13Grok 4.757.4%
  4. #14Gemini 3.8 Flash56.6%
artificialanalysis.ai · 2026-09-24
~#17/103~56.1%Estimated
  1. #1Claude Opus 5.566.9%
  2. #16Claude Opus 556.4%
  3. ~#17Ternary Bonsai 2 27BEstimated56.1%
  4. #17DeepSeek V4.1 Flash NVFP455.8%
estimated from related benchmarks
AA Coding Agent Index
third-party
~#7/15~54.4Estimated
  1. #1Claude Opus 5.566.0
  2. #6Grok 4.756.3
  3. ~#7GLM-5.3 Flash NVFP4Estimated54.4
  4. #7Muse Spark 1.354.3
estimated from related benchmarks
#6/1456.3Verified
  1. #1Claude Opus 5.566.0
  2. #5GPT-6 Sol56.7
  3. #6Grok 4.756.3
  4. #7Muse Spark 1.354.3
artificialanalysis.ai · 2026-09-24
~#5/15~59.5Estimated
  1. #1Claude Opus 5.566.0
  2. #4Claude Opus 559.7
  3. ~#5Ternary Bonsai 2 27BEstimated59.5
  4. #5GPT-6 Sol56.7
estimated from related benchmarks
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
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
IFBench
day 0
#50/13360.5%Lab-reported
  1. #1MiniMax-M382.9%
  2. #49K2 Think V262.8%
  3. #50GLM-5.3 Flash NVFP460.5%
  4. #51K2-V260.1%
huggingface.co · 2026-09-29
~#32/134~70.7%Estimated
  1. #1MiniMax-M382.9%
  2. #31Nova 2.0 Lite70.7%
  3. ~#32Grok 4.7Estimated70.7%
  4. #32Xing4.0-29B-A4B69.7%
estimated from related benchmarks
#23/13374.0%BestLab-reported
  1. #1MiniMax-M382.9%
  2. #22GPT-5.3 Codex75.4%
  3. #23Ternary Bonsai 2 27B74.0%
  4. #24Command A+73.9%
huggingface.co · 2026-09-29
MMMLU
day 0
~#3/5~86.7%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. #2Gemma 4 31B88.4%
  3. ~#3GLM-5.3 Flash NVFP4Estimated86.7%
  4. #3Claude Haiku 4.583.0%
estimated from related benchmarks
~#2/5~89.2%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. ~#2Grok 4.7Estimated89.2%
  3. #2Gemma 4 31B88.4%
estimated from related benchmarks
~#2/5~89.2%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. ~#2Ternary Bonsai 2 27BEstimated89.2%
  3. #2Gemma 4 31B88.4%
estimated from related benchmarks
LMArena Text, Non-English (Elo)
third-party
~#23/54~1440Estimated
  1. #1Claude Opus 5.51501
  2. #22Hy31440
  3. ~#23GLM-5.3 Flash NVFP4Estimated1440
  4. #23GPT-5.4 mini1439
estimated from related benchmarks
#26/531435Verified
  1. #1Claude Opus 5.51501
  2. #25GPT-6 Luna1436
  3. #26Grok 4.71435
  4. #27Qwen3.5 397B A17B1434
lmarena.ai · 2026-09-25
~#16/54~1454Estimated
  1. #1Claude Opus 5.51501
  2. #15GPT-5.6 Terra1456
  3. ~#16Ternary Bonsai 2 27BEstimated1454
  4. #16DeepSeek V4 Pro (0813)1448
estimated from related benchmarks
LMArena Text (Elo)
third-party
~#23/54~1452Estimated
  1. #1Claude Opus 5.51509
  2. #22Grok 4.61453
  3. ~#23GLM-5.3 Flash NVFP4Estimated1452
  4. #23Gemma 4 31B1451
estimated from related benchmarks
#29/531439Verified
  1. #1Claude Opus 5.51509
  2. #28Inkling1440
  3. #29Grok 4.71439
  4. #30Gemma 4 26B A4B1438
lmarena.ai · 2026-09-25
~#16/54~1465Estimated
  1. #1Claude Opus 5.51509
  2. #15GPT-5.6 Terra1466
  3. ~#16Ternary Bonsai 2 27BEstimated1465
  4. #16DeepSeek V4 Pro (0813)1463
estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)
third-party
~#9/26~1966Estimated
  1. #1GPT-6 Astra2173
  2. #8Grok 4.72007
  3. ~#9GLM-5.3 Flash NVFP4Estimated1966
  4. #9Muse Spark 1.31906
estimated from related benchmarks
#8/252007Verified
  1. #1GPT-6 Astra2173
  2. #7Claude Opus 5.52050
  3. #8Grok 4.72007
  4. #9Muse Spark 1.31906
eqbench.com · 2026-09-24
~#9/26~1983Estimated
  1. #1GPT-6 Astra2173
  2. #8Grok 4.72007
  3. ~#9Ternary Bonsai 2 27BEstimated1983
  4. #9Muse Spark 1.31906
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
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
MMMU
day 0
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2GLM-5.3 Flash NVFP4Estimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2Grok 4.7Estimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2Ternary Bonsai 2 27BEstimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
MMMU-Pro
day 0
#28/8476.3%BestLab-reported
  1. #1Claude Opus 5.587.7%
  2. #27Step 5 Preview76.4%
  3. #28GLM-5.3 Flash NVFP476.3%
  4. #28Qwen3.8 27B76.3%
huggingface.co · 2026-09-29
~#21/85~78.5%Estimated
  1. #1Claude Opus 5.587.7%
  2. #20GPT-5.3 Codex78.5%
  3. ~#21Grok 4.7Estimated78.5%
  4. #21Qwen3.8 Flash-Next NVFP478.3%
estimated from related benchmarks
#31/8475.5%Lab-reported
  1. #1Claude Opus 5.587.7%
  2. #30GPT-6 Luna75.5%
  3. #31Ternary Bonsai 2 27B75.5%
  4. #32MiMo-V2.575.4%
huggingface.co · 2026-09-29
CharXiv Reasoning
day 0
~#5/9~81.2%Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #4Qwen3.8-Omni-Flash83.5%
  3. ~#5GLM-5.3 Flash NVFP4Estimated81.2%
  4. #5Muse Glimmer78.8%
estimated from related benchmarks
~#5/9~81.4%Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #4Qwen3.8-Omni-Flash83.5%
  3. ~#5Grok 4.7Estimated81.4%
  4. #5Muse Glimmer78.8%
estimated from related benchmarks
~#5/9~81.2%Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #4Qwen3.8-Omni-Flash83.5%
  3. ~#5Ternary Bonsai 2 27BEstimated81.2%
  4. #5Muse Glimmer78.8%
estimated from related benchmarks
OmniDocBench v1.5
third-party
~#4/8~87.6%Estimated
  1. #1MiniMax-M391.6%
  2. #2Qwen3.8 27B91.1%
  3. ~#4GLM-5.3 Flash NVFP4Estimated87.6%
  4. #4Gemini 3.1 Pro85.3%
estimated from related benchmarks
~#4/8~87.1%Estimated
  1. #1MiniMax-M391.6%
  2. #2Qwen3.8 27B91.1%
  3. ~#4Grok 4.7Estimated87.1%
  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. ~#4Ternary Bonsai 2 27BEstimated87.3%
  4. #4Gemini 3.1 Pro85.3%
estimated from related benchmarks
GDP.pdf (AA)
third-party
~#17/43~16.9%Estimated
  1. #1GPT-6 Astra31.0%
  2. #16Grok 4.617.2%
  3. ~#17GLM-5.3 Flash NVFP4Estimated16.9%
  4. #17Qwen3.8 27B16.6%
estimated from related benchmarks
#13/4220.0%Verified
  1. #1GPT-6 Astra31.0%
  2. #12GPT-6 Luna20.4%
  3. #13Grok 4.720.0%
  4. #14MiMo-V2.6-Pro19.2%
artificialanalysis.ai · 2026-09-24
~#16/43~17.8%Estimated
  1. #1GPT-6 Astra31.0%
  2. #15Gemini 3.1 Pro17.8%
  3. ~#16Ternary Bonsai 2 27BEstimated17.8%
  4. #16Grok 4.617.2%
estimated from related benchmarks
Chartography
third-party
~#10/23~28.6%Estimated
  1. #1GPT-6 Astra71.0%
  2. #8Qwen3.8 Max (0902)29.1%
  3. ~#10GLM-5.3 Flash NVFP4Estimated28.6%
  4. #10Muse Spark 1.327.6%
estimated from related benchmarks
#18/2214.7%Verified
  1. #1GPT-6 Astra71.0%
  2. #17GLM-5.3 Flash16.3%
  3. #18Grok 4.714.7%
  4. #19DeepSeek V4.1 Flash11.6%
surgehq.ai · 2026-09-24
~#8/23~29.7%Estimated
  1. #1GPT-6 Astra71.0%
  2. #7GPT-5.531.0%
  3. ~#8Ternary Bonsai 2 27BEstimated29.7%
  4. #8GPT-6 Luna29.1%
estimated from related benchmarks

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
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
GPQA Diamond
day 0
#19/17992.1%Lab-reported
  1. #1GPT-6 Astra96.1%
  2. #17Qwen3.8 Flash-Next92.3%
  3. #19GLM-5.3 Flash NVFP492.1%
  4. #20GLM-5.391.7%
huggingface.co · 2026-09-29
~#20/180~92.0%Estimated
  1. #1GPT-6 Astra96.1%
  2. #19GLM-5.3 Flash NVFP492.1%
  3. ~#20Grok 4.7Estimated92.0%
  4. #20GLM-5.391.7%
estimated from related benchmarks
~#37/180~88.5%Estimated
  1. #1GPT-6 Astra96.1%
  2. #36Solar Pro 489.1%
  3. ~#37Ternary Bonsai 2 27BEstimated88.5%
  4. #37Qwen3.8 27B NVFP488.0%
estimated from related benchmarks
Humanity's Last Exam
day 0
~#44/182~31.1%Estimated
  1. #1Claude Opus 5.561.4%
  2. #43Inkling31.9%
  3. ~#44GLM-5.3 Flash NVFP4Estimated31.1%
  4. #44A.X-K229.6%
estimated from related benchmarks
#17/18143.1%Verified
  1. #1Claude Opus 5.561.4%
  2. #16Hy4 preview43.4%
  3. #17Grok 4.743.1%
  4. #17Qwen3.8 Max (0902)43.1%
artificialanalysis.ai · 2026-09-24
~#37/182~34.0%Estimated
  1. #1Claude Opus 5.561.4%
  2. #36Apodex 1.134.1%
  3. ~#37Ternary Bonsai 2 27BEstimated34.0%
  4. #37Qwen3.8 27B33.9%
estimated from related benchmarks
AIME (latest)
day 0
~#2/30~98.9%Estimated
  1. #1GLM-5.3 NVFP499.5%
  2. ~#2GLM-5.3 Flash NVFP4Estimated98.9%
  3. #2Inkling97.1%
estimated from related benchmarks
~#4/30~95.0%Estimated
  1. #1GLM-5.3 NVFP499.5%
  2. #3Ternary Bonsai 2 27B95.8%
  3. ~#4Grok 4.7Estimated95.0%
  4. #4Muse Glimmer94.7%
estimated from related benchmarks
#3/2995.8%Lab-reported
  1. #1GLM-5.3 NVFP499.5%
  2. #2Inkling97.1%
  3. #3Ternary Bonsai 2 27B95.8%
  4. #4Muse Glimmer94.7%
huggingface.co · 2026-09-29
AA Intelligence Index
third-party
~#26/78~36.0Estimated
  1. #1Claude Opus 5.557.6
  2. #25DeepSeek V4 Pro (0813)36.0
  3. ~#26GLM-5.3 Flash NVFP4Estimated36.0
  4. #26Qwen3.8 27B33.7
estimated from related benchmarks
#9/7746.4Verified
  1. #1Claude Opus 5.557.6
  2. #8GPT-6 Sol47.5
  3. #9Grok 4.746.4
  4. #10MiMo-V2.6-Pro46.3
artificialanalysis.ai · 2026-09-24
~#22/78~39.1Estimated
  1. #1Claude Opus 5.557.6
  2. #21DeepSeek V4.1 Flash39.5
  3. ~#22Ternary Bonsai 2 27BEstimated39.1
  4. #22GPT-5.538.4
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
GLM-5.3 Flash NVFP4
Grok 4.7
Ternary Bonsai 2 27B
τ²-bench (Telecom)
day 0
—
~#4/117~95.6%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #2Qwen3.5 397B A17B95.6%
  3. ~#4Grok 4.7Estimated95.6%
  4. #4Qwen3.6 35B A3B95.3%
estimated from related benchmarks
—
OSWorld-Verified
day 0
—
~#3/12~82.3%Estimated
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. ~#3Grok 4.7Estimated82.3%
  4. #3Nex-N2.5-Pro82.2%
estimated from related benchmarks
—
OSWorld 2.0
day 0
—
~#3/5~58.3%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Grok 4.7Estimated58.3%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
—
BrowseComp
day 0
—
~#7/18~88.5%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #6Step 5 Preview88.7%
  3. ~#7Grok 4.7Estimated88.5%
  4. #7GPT-5.6 Terra87.5%
estimated from related benchmarks
—
GDPval-AA (Elo)
third-party
—
#4/861695Verified
  1. #1Claude Opus 5.51846
  2. #3Claude Opus 51708
  3. #4Grok 4.71695
  4. #5Muse Spark 1.31674
artificialanalysis.ai · 2026-09-24
—
AutomationBench-AA
third-party
—
#5/7965.6%Verified
  1. #1Claude Opus 5.569.5%
  2. #4Grok 4.667.0%
  3. #5Grok 4.765.6%
  4. #6GLM-5.362.2%
artificialanalysis.ai · 2026-09-24
—
τ-Bench Banking (AA)
third-party
—
~#11/90~43.3%Estimated
  1. #1Muse Spark 1.350.5%
  2. #10Gemini 3.8 Flash44.9%
  3. ~#11Grok 4.7Estimated43.3%
  4. #11Grok 4.643.3%
estimated from related benchmarks
—