Inkling Small
- Input
- $0.30/M tok
- Output
- $1.20/M tok
- Blended
- $0.525/M tok · 3:1
- Speed
- 206tok/s
- Context
- 1Mtokens
- Cost per task
- ~$0.51/taskest.
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Inkling Small. Price = AA median host price.
Rankings
Coding
Verified#66 / 137rank by quality44.4±8.1quality / 1002/7 benchmarks measuredVision
Verified#47 / 98rank by quality56.9±8.2quality / 1001/6 benchmarks measuredHard reasoning
Verified#65 / 218rank by quality60.9±3.7quality / 1003/4 benchmarks measuredAgents
Verified#61 / 187rank by quality43.2±10.0quality / 1001/7 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
- 4× B200 180GB
- FP8 weights
- Throughput
- ~22,891 tok/s
- many requests batched
- Price
- $0.12–1.03 /M tok
- busy → light use
- On your own machine
- Mac Studio M5 Ultra 512GB (int4, ~140 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (276B params, 12B active per token) + 40% KV-cache headroom ≈ 386 GB
- 4× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~22,891 output tok/s aggregate at batch 655 (bandwidth-bound); MoE compute scales with active params
- Blended 3:1 input:output; prefill ~354,375 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 VerifiedEstimated
~#9 of 21 (estimated)~76.9%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #7Mistral Medium 3.577.6%
- ~#9Inkling SmallEstimated76.9%
- #9Ripple-65B Preview76.5%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#17 of 24 (estimated)~56.7%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #16Hy357.9%
- ~#17Inkling SmallEstimated56.7%
- #17Inkling54.3%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#37 of 46 (estimated)~52.6%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #36Nex-N2.5-Pro55.8%
- ~#37Inkling SmallEstimated52.6%
- #37Qwen3.8 Max (0902)51.0%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#15 of 28 (estimated)~73.4%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #14Maple-Preview75.1%
- ~#15Inkling SmallEstimated73.4%
- #15Mistral Small 464.0%
No published score — estimated from related benchmarks
SciCodeVerified
#45 of 11849.7%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #44Qwen3.8 Flash-Next50.6%
- #45Inkling Small49.7%
- #46Beam 1.0 501B A23B49.2%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#23 of 26 (estimated)~33.9
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #22Claude Haiku 5.537.3
- ~#23Inkling SmallEstimated33.9
- #23Gemini 3.8 Flash33.3
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#41 of 8774.0%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #40Muse Glimmer74.3%
- #41Inkling Small74.0%
- #42DeepSeek V4.1 Flash NVFP473.7%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~#10 of 12 (estimated)~77.9%
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #9Inkling78.1%
- ~#10Inkling SmallEstimated77.9%
- #10Claude Sonnet 577.0%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#5 of 8 (estimated)~84.1%
- #1MiniMax-M391.6%
- #2Kimi K391.1%
- #4Gemini 3.1 Pro85.3%
- ~#5Inkling SmallEstimated84.1%
- #5Claude Haiku 4.579.6%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#29 of 48 (estimated)~12.9%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #28Claude Sonnet 513.2%
- ~#29Inkling SmallEstimated12.9%
- #29DeepSeek V4.1 Flash12.8%
No published score — estimated from related benchmarks
ChartographyEstimated
~#17 of 25 (estimated)~18.8%
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #16Grok 4.620.5%
- ~#17Inkling SmallEstimated18.8%
- #17Muse Glimmer17.2%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#36 of 19289.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #36Grok Build 0.1 061689.5%
- #36Inkling Small89.5%
- #38Qwen3.5 397B A17B89.3%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#52 of 19833.3%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #51Hy333.5%
- #52Inkling Small33.3%
- #53K2 Horizon 375B A23B32.0%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#4 of 34 (estimated)~96.8%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #3Inkling97.1%
- ~#4Inkling SmallEstimated96.8%
- #4Ternary Bonsai 2 27B95.8%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Estimated
~#37 of 120 (estimated)~78.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #35Qwen3 Coder Next79.5%
- ~#37Inkling SmallEstimated78.8%
- #37EXAONE 4.5 33B78.1%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#11 of 15 (estimated)~67.6%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- ~#11Inkling SmallEstimated67.6%
- #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
~#19 of 22 (estimated)~67.7%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Inkling SmallEstimated67.7%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#29 of 92 (estimated)~1380 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #28Qwen3.8 27B1419
- ~#29Inkling SmallEstimated1380
- #29K2 Horizon 375B A23B1361
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#38 of 85 (estimated)~23.9%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #37Kimi K2.7 Code24.5%
- ~#38Inkling SmallEstimated23.9%
- #38Quasar 438B23.9%
No published score — estimated from related benchmarks