K2 Horizon MoVA 36B A4B
- Input
- —
- Output
- —
- Speed
- ~120tok/sest.
- Context
- 524Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: K2 Horizon MoVA 36B A4B. Price = AA median host price.
Rankings
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.
- Hardware
- 1× H200 141GB
- FP8 weights
- Throughput
- ~7,534 tok/s
- many requests batched
- Price
- $0.053–0.44 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, 300+ tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (36B params, 4B active per token) + 40% KV-cache headroom ≈ 50 GB
- 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
- ~7,534 output tok/s aggregate at batch 256 (bandwidth-bound); MoE compute scales with active params
- Blended 3:1 input:output; prefill ~116,885 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
~#16 of 21 (estimated)~71.0%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #15Nemotron 3.5 Super VL NVFP471.1%
- ~#16K2 Horizon MoVA 36B A4BEstimated71.0%
- #16Nemotron 3 Ultra70.7%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~50.1%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23K2 Horizon MoVA 36B A4BEstimated50.1%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#38 of 46 (estimated)~44.6%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #37Qwen3.8 Max (0902)51.0%
- ~#38K2 Horizon MoVA 36B A4BEstimated44.6%
- #38Ripple-65B Preview44.3%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#63 of 1010.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #63K2 Horizon 3.7B0.0%
- #63K2 Horizon MoVA 36B A4B0.0%
- #63Ling 3.0 Flash0.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchEstimated
~#16 of 28 (estimated)~63.0%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #15Mistral Small 464.0%
- ~#16K2 Horizon MoVA 36B A4BEstimated63.0%
- #16Claude Haiku 4.561.5%
No published score — estimated from related benchmarks
SciCodeVerified
#80 of 11840.0%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #80Gemma 4 26B A4B40.0%
- #80K2 Horizon MoVA 36B A4B40.0%
- #82Nemotron 3.5 Bolt 74B-A9B39.7%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~30.2
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25K2 Horizon MoVA 36B A4BEstimated30.2
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#72 of 19282.2%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #70K2-Horizon-32B82.3%
- #72K2 Horizon MoVA 36B A4B82.2%
- #73GPT-5.4 nano81.7%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#70 of 19823.4%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #69Gemma 4 31B23.6%
- #70K2 Horizon MoVA 36B A4B23.4%
- #71K2-Horizon-32B22.8%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#15 of 34 (estimated)~87.6%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #14Nova 2.0 Pro Preview89.0%
- ~#15K2 Horizon MoVA 36B A4BEstimated87.6%
- #15Maple-Preview87.5%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Estimated
~#42 of 120 (estimated)~70.5%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #41Nova 2.0 Lite72.8%
- ~#42K2 Horizon MoVA 36B A4BEstimated70.5%
- #42Apriel-v1.6-15B-Thinker69.3%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#11 of 15 (estimated)~70.8%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- ~#11K2 Horizon MoVA 36B A4BEstimated70.8%
- #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)~73.0%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #17Beam 1.0 501B A23B76.8%
- ~#18K2 Horizon MoVA 36B A4BEstimated73.0%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#33 of 911181 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #32Apodex 1.11189
- #33K2 Horizon MoVA 36B A4B1181
- #34Quasar 438B1162
artificialanalysis.ai · 2026-09-24