Muse Glimmer
- Input
- $0.325/M tok
- Output
- $1.35/M tok
- Blended
- $0.581/M tok · 3:1
- Speed
- 134tok/s
- Context
- 131Ktokens
- Cost per task
- ~$0.57/taskest.
Open weights (30B, Apache 2.0); no first-party API, price is the Artificial Analysis third-party host price.
Rankings
Coding
Mixed#84 / 137rank by quality38.5±5.4quality / 1004/7 benchmarks measuredWriting & chat
Mixed#56 / 162rank by quality59.5±4.4quality / 1004/5 benchmarks measuredVision
Mixed#59 / 98rank by quality48.0±3.8quality / 1005/6 benchmarks measuredHard reasoning
Mixed#89 / 218rank by quality51.5±3.6quality / 1004/4 benchmarks measuredAgents
Mixed#88 / 187rank by quality35.8±4.7quality / 1004/7 benchmarks measured
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.
- Hardware
- 1× B200 180GB
- BF16 weights
- Throughput
- ~4,025 tok/s
- many requests batched
- Price
- $0.23–2.05 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, ~85 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- BF16 weights (29.6B params) + 25% KV-cache headroom ≈ 74 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~4,025 output tok/s aggregate at batch 115 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~17,958 tok/s
- Low = 75% utilization at the low GPU price; high = 20% at the high price
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedLab-reported
#10 of 2076.0%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #9Ripple-65B Preview76.5%
- #10Muse Glimmer76.0%
- #11Xing4.0-29B-A4B75.0%
huggingface.co · 2026-08-10
SWE-bench Pro (Public)Lab-reported
#22 of 2351.2%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #21Nemotron 3.5 Super VL NVFP451.3%
- #22Muse Glimmer51.2%
- #23Nex-N2.5-Mini43.8%
huggingface.co · 2026-08-10
DeepSWE v1.1Estimated
~#38 of 46 (estimated)~45.6%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #37Qwen3.8 Max (0902)51.0%
- ~#38Muse GlimmerEstimated45.6%
- #38Ripple-65B Preview44.3%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#15 of 28 (estimated)~66.6%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #14Maple-Preview75.1%
- ~#15Muse GlimmerEstimated66.6%
- #15Mistral Small 464.0%
No published score — estimated from related benchmarks
SciCodeVerified
#63 of 11844.9%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #62Ring-2.6-1T45.0%
- #63Muse Glimmer44.9%
- #64Qwen3.5 397B A17B44.8%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#24 of 26 (estimated)~32.5
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #23Gemini 3.8 Flash33.3
- ~#24Muse GlimmerEstimated32.5
- #24GPT-6 Luna31.1
No published score — estimated from related benchmarks
Writing & chat
IFBenchLab-reported
#19 of 13977.0%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #18Gemini 3.1 Pro77.1%
- #19Muse Glimmer77.0%
- #20GPT-5.4 nano75.9%
huggingface.co · 2026-08-10
LMArena Text, Non-English (Elo)Few votesVerified
#35 of 561418 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #34Gemma 4 26B A4BFew votes1423
- #35Muse GlimmerFew votes1418
- #36MiMo-V2.51411
lmarena.ai · 2026-09-25
LMArena Text (Elo)Few votesVerified
#38 of 561424 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #36Nemotron 3 Ultra1426
- #38Muse GlimmerFew votes1424
- #39Qwen3.5 122B A10B1417
lmarena.ai · 2026-09-25
#13 of 251798 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #12Qwen3.8 2.4T A95B1843
- #13Muse Glimmer1798
- #14Claude Sonnet 51794
eqbench.com · 2026-09-24
Vision
MMMU-ProVerified
#40 of 8774.3%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #39Qwen3.8 27B NVFP474.9%
- #40Muse Glimmer74.3%
- #41Inkling Small74.0%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningLab-reported
#8 of 1178.8%
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #7Nemotron 3.5 Super VL NVFP479.9%
- #8Muse Glimmer78.8%
- #9Inkling78.1%
huggingface.co · 2026-08-10
ChartographyVerified
#17 of 2417.2%
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #16Grok 4.620.5%
- #17Muse Glimmer17.2%
- #18Claude Sonnet 516.6%
surgehq.ai · 2026-09-24
Hard reasoning
GPQA DiamondVerified
#65 of 19283.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #64Gemini 3.5 Flash-Lite83.8%
- #65Muse Glimmer83.5%
- #66Motif 383.4%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#74 of 19822.0%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #74Ling-3.0-flash-VL22.0%
- #74Muse Glimmer22.0%
- #76Intern-S2-Preview (35B-A3B)21.9%
artificialanalysis.ai · 2026-09-24
AIME (latest)Lab-reported
#6 of 3394.7%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #5Ripple-65B Preview95.0%
- #6Muse Glimmer94.7%
- #7Nova 2.0 Lite94.3%
huggingface.co · 2026-08-10
Agents
τ²-bench (Telecom)Estimated
~#42 of 120 (estimated)~69.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #41Nova 2.0 Lite72.8%
- ~#42Muse GlimmerEstimated69.8%
- #42Apriel-v1.6-15B-Thinker69.3%
No published score — estimated from related benchmarks
OSWorld-VerifiedLab-reported
#11 of 1465.9%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- #11Muse Glimmer65.9%
- #12Nemotron 3.5 Super VL NVFP452.3%
huggingface.co · 2026-08-10
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#19 of 22 (estimated)~68.3%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Muse GlimmerEstimated68.3%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#52 of 91786 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #51Qwen3.5 122B A10B802
- #52Muse Glimmer786
- #53Qwen3.5 397B A17B780
artificialanalysis.ai · 2026-10-10
AutomationBench-AAVerified
#48 of 846.8%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #47Qwen3.5 397B A17B6.9%
- #48Muse Glimmer6.8%
- #49Mistral Medium 3.56.3%
artificialanalysis.ai · 2026-09-24