Mercury 2.5
- Input
- $0.20/M tok
- Output
- $0.75/M tok
- Blended
- $0.338/M tok · 3:1
- Speed
- 781tok/s
- Context
- 260Ktokens
- Cost per task
- ~$0.32/taskest.
Diffusion LLM; vendor claims 1,107 tok/s. Launch discount $0.04/$0.15 (80% off).
Rankings
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#17 of 21 (estimated)~68.8%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #16Nemotron 3 Ultra70.7%
- ~#17Mercury 2.5Estimated68.8%
- #17Intern-S2-Preview (35B-A3B)64.0%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~47.9%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23Mercury 2.5Estimated47.9%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#40 of 46 (estimated)~42.2%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #39Beam 1.0 501B A23B43.3%
- ~#40Mercury 2.5Estimated42.2%
- #40Qwen3.8 27B42.2%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#17 of 28 (estimated)~55.5%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #16Claude Haiku 4.561.5%
- ~#17Mercury 2.5Estimated55.5%
- #17GLM-5.3 NVFP453.5%
No published score — estimated from related benchmarks
SciCodeVerified
#88 of 11838.5%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #88Command A+38.5%
- #88Mercury 2.538.5%
- #90Granite 4.2 30B37.8%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~28.6
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25Mercury 2.5Estimated28.6
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondEstimated
~#129 of 193 (estimated)~63.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #128LongCat Flash Lite63.6%
- ~#129Mercury 2.5Estimated63.5%
- #129LFM3-1.7B63.4%
No published score — estimated from related benchmarks
Humanity's Last ExamVerified
#101 of 19811.8%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #98Nemotron Cascade 2 30B A3B12.0%
- #101Mercury 2.511.8%
- #102Nova 2.0 Lite11.6%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~59.2%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Mercury 2.5Estimated59.2%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Estimated
~#59 of 120 (estimated)~33.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #58Sarvam 30B34.5%
- ~#59Mercury 2.5Estimated33.8%
- #59Gemma 4 12B31.9%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~60.5%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Mercury 2.5Estimated60.5%
- #12Nemotron 3.5 Super VL NVFP452.3%
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
~#19 of 22 (estimated)~56.6%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Mercury 2.5Estimated56.6%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#73 of 91462 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #72Nemotron 3 Super 120B A12B477
- #73Mercury 2.5462
- #74EXAONE 4.5 33B458
artificialanalysis.ai · 2026-09-24
AutomationBench-AAVerified
#61 of 842.4%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #60Nemotron 3 Ultra3.0%
- #61Mercury 2.52.4%
- #62Granite 4.2 8B2.1%
artificialanalysis.ai · 2026-09-24
τ-Bench Banking (AA)Estimated
~#63 of 91 (estimated)~10.4%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #62Devstral 210.5%
- ~#63Mercury 2.5Estimated10.4%
- #63Nemotron 3 Super 120B A12B10.3%
No published score — estimated from related benchmarks