Skip to content
Model Pareto

K2 Horizon MoVA 36B A4B

MBZUAI Institute of Foundation ModelsOpen weightsartificialanalysis.ai ↗
Compare with nearest rivalsCompare with…
Input
—
Output
—
Speed
~110tok/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

DeepSWE v1.1Estimated
~#28 of 31 (estimated)~48.0%
  1. #1DeepSeek V4.1 Flash74.2%
  2. #2Grok 4.772.6%
  3. #27Qwen3.8 Max (0902)51.0%
  4. ~#28K2 Horizon MoVA 36B A4BEstimated48.0%
  5. #28Qwen3.8 27B42.2%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~last of 15 (estimated)below measured range (<41.1)
  1. #1Claude Opus 5.566.0
  2. #2Claude Fable 5.162.2
  3. #14GPT-6 Luna41.1
  4. ~#15K2 Horizon MoVA 36B A4BEstimated<41.1
No published score — estimated from related benchmarks

Hard reasoning

AIME (latest)Estimated
~#5 of 25 (estimated)~90.9%
  1. #1Inkling97.1%
  2. #2Muse Glimmer94.7%
  3. #4gpt-oss-120b93.4%
  4. ~#5K2 Horizon MoVA 36B A4BEstimated90.9%
  5. #5Xing4.0-29B-A4B90.0%
No published score — estimated from related benchmarks

Agents

OSWorld-VerifiedEstimated
~#9 of 12 (estimated)~71.5%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #8Gemini 3.5 Flash-Lite74.0%
  4. ~#9K2 Horizon MoVA 36B A4BEstimated71.5%
  5. #9Nex-N2.5-Mini71.2%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #4GPT-5.547.5%
  4. ~#5K2 Horizon MoVA 36B A4BEstimated<47.5%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#15 of 18 (estimated)~72.6%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #14Inkling77.1%
  4. ~#15K2 Horizon MoVA 36B A4BEstimated72.6%
  5. #15Mistral Medium 3.548.6%
No published score — estimated from related benchmarks