Quasar 438B
- Input
- $0.60/M tok
- Output
- $1.80/M tok
- Blended
- $0.90/M tok · 3:1
- Speed
- 152tok/s
- Context
- 1Mtokens
- Cost per task
- ~$0.76/taskest.
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Quasar 438B (max, based on GLM-5.2). Price = AA median host price.
Rankings
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#12 of 21 (estimated)~74.9%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #11Xing4.0-29B-A4B75.0%
- ~#12Quasar 438BEstimated74.9%
- #12Flint74.2%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#20 of 24 (estimated)~53.9%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #18Gemini 3.5 Flash-Lite54.2%
- ~#20Quasar 438BEstimated53.9%
- #20Nemotron 3.5 Super VL51.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#38 of 46 (estimated)~50.6%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #37Qwen3.8 Max (0902)51.0%
- ~#38Quasar 438BEstimated50.6%
- #38Ripple-65B Preview44.3%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#46 of 1011.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #46Kimi K2.7 Code1.0%
- #46Quasar 438B1.0%
- #46Qwen3.7 Plus1.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchEstimated
~#15 of 28 (estimated)~68.0%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #14Maple-Preview75.1%
- ~#15Quasar 438BEstimated68.0%
- #15Mistral Small 464.0%
No published score — estimated from related benchmarks
SciCodeVerified
#49 of 11848.1%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #48Qwen3.8 27B NVFP448.4%
- #49Quasar 438B48.1%
- #50Solar Open2 250B48.0%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#24 of 26 (estimated)~31.8
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #23Gemini 3.8 Flash33.3
- ~#24Quasar 438BEstimated31.8
- #24GPT-6 Luna31.1
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#106 of 19273.2%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #104HyperNova 60B 260573.3%
- #106Quasar 438B73.2%
- #107Llama 3.1 Nemotron Ultra 253B v172.8%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#85 of 19818.7%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #84Gemini 3.5 Flash-Lite18.8%
- #85Quasar 438B18.7%
- #86K-EXAONE 2.0 080318.6%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~77.5%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Quasar 438BEstimated77.5%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Estimated
~#44 of 120 (estimated)~66.7%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #43Nemotron 3 Super 120B A12B67.8%
- ~#44Quasar 438BEstimated66.7%
- #44gpt-oss-120b65.8%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#11 of 15 (estimated)~69.4%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- ~#11Quasar 438BEstimated69.4%
- #11Muse Glimmer65.9%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#18 of 22 (estimated)~70.7%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #17Beam 1.0 501B A23B76.8%
- ~#18Quasar 438BEstimated70.7%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#34 of 911162 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #33K2 Horizon MoVA 36B A4B1181
- #34Quasar 438B1162
- #35Ling-3.0-flash-VL1150
artificialanalysis.ai · 2026-09-24
AutomationBench-AAVerified
#38 of 8423.9%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #37Kimi K2.7 Code24.5%
- #38Quasar 438B23.9%
- #39MiniMax-M321.3%
artificialanalysis.ai · 2026-09-24