Skip to content
Model Pareto

Hy3

Tencent HunyuanOpen weightshuggingface.co ↗
Compare with nearest rivalsCompare with…
Input
$0.14/M tok
Output
$0.58/M tok
Blended
$0.25/M tok · 3:1
Speed
86tok/s
Context
256Ktokens

Open weights.

Rankings

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
~16,423 tok/s
many requests batched
Price
$0.17–1.50 /M tok
busy → light use
On your own machine
Mac Studio M5 Ultra 512GB (int4, ~80 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (295B params, 21B active per token) + 25% KV-cache headroom ≈ 369 GB
  • 4× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~16,423 output tok/s aggregate at batch 470 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~202,500 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

LiveCodeBenchEstimated
~#7 of 20 (estimated)~78.0%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #6Gemma 4 31B80.0%
  4. ~#7Hy3Estimated78.0%
  5. #7gpt-oss-20b77.7%
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. ~#15Hy3Estimated<41.1
No published score — estimated from related benchmarks

Writing & chat

IFBenchEstimated
~#33 of 126 (estimated)~67.2%
  1. #1MiniMax-M382.9%
  2. #2Qwen3.8 2.4T A95B82.8%
  3. #32Step 3.7 Flash67.3%
  4. ~#33Hy3Estimated67.2%
  5. #33MiMo-V2.567.1%
No published score — estimated from related benchmarks
MMMLUEstimated
~#3 of 5 (estimated)~87.8%
  1. #1Gemini 3.1 Pro92.6%
  2. #2Gemma 4 31B88.4%
  3. ~#3Hy3Estimated87.8%
  4. #3Claude Haiku 4.583.0%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Verified
#22 of 531440 Elo
  1. #1Claude Opus 5.51501
  2. #2Claude Fable 5.1Few votes1496
  3. #21Qwen3.7 Plus1443
  4. #22Hy31440
  5. #23GPT-5.4 mini1439
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#19 of 531456 Elo
  1. #1Claude Opus 5.51509
  2. #2Claude Fable 5.11501
  3. #19Gemini 3.5 Flash-Lite1456
  4. #19Hy31456
  5. #19Qwen3.7 Plus1456
lmarena.ai · 2026-09-24
EQ-Bench Creative Writing v3 (Elo)Estimated
~#13 of 26 (estimated)~1832 Elo
  1. #1GPT-6 Astra2173
  2. #2Claude Fable 5.12162
  3. #12Qwen3.8 2.4T A95B1843
  4. ~#13Hy3Estimated1832
  5. #13Muse Glimmer1798
No published score — estimated from related benchmarks

Hard reasoning

AIME (latest)Estimated
~#5 of 25 (estimated)~92.8%
  1. #1Inkling97.1%
  2. #2Muse Glimmer94.7%
  3. #4gpt-oss-120b93.4%
  4. ~#5Hy3Estimated92.8%
  5. #5Xing4.0-29B-A4B90.0%
No published score — estimated from related benchmarks

Agents

τ²-bench (Telecom)Estimated
~#36 of 116 (estimated)~77.2%
  1. #1Step 3.7 Flash98.5%
  2. #2Gemini 3.1 Pro95.6%
  3. #35EXAONE 4.5 33B78.1%
  4. ~#36Hy3Estimated77.2%
  5. #36GPT-5.4 nano76.0%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#10 of 12 (estimated)~70.1%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #9Nex-N2.5-Mini71.2%
  4. ~#10Hy3Estimated70.1%
  5. #10Muse Glimmer65.9%
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. ~#5Hy3Estimated<47.5%
No published score — estimated from related benchmarks