Hermes 4 - Llama-3.1 405B
- Input
- $1.00/M tok
- Output
- $3.00/M tok
- Blended
- $1.50/M tok · 3:1
- Speed
- 42tok/s
- Context
- 128Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Hermes 4 - Llama-3.1 405B (Reasoning). Price = AA median host price.
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
- ~2,555 tok/s
- many requests batched
- Price
- $1.50–13.4 /M tok
- busy → light use
- On your own machine
- Mac Studio M5 Ultra 512GB (int4, ~4 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (405.9B params) + 25% KV-cache headroom ≈ 507 GB
- 4× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~2,555 output tok/s aggregate at batch 73 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~10,477 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.
Writing & chat
IFBenchVerified
#108 of 13932.7%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #107Olmo 3 7B Instruct32.8%
- #108Hermes 4 - Llama-3.1 405B32.7%
- #109Ministral 3 14B32.0%
artificialanalysis.ai · 2026-09-24
MMMLUEstimated
~last of 5 (estimated)below measured range (<81.3%)
- #1Gemini 3.1 Pro92.6%
- #2Gemma 4 31B88.4%
- #4gpt-oss-120b81.3%
- ~#5Hermes 4 - Llama-3.1 405BEstimated<81.3%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Estimated
~#47 of 57 (estimated)~1331 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #46gpt-oss-120b1338
- ~#47Hermes 4 - Llama-3.1 405BEstimated1331
- #47INTELLECT-3Few votes1327
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#47 of 57 (estimated)~1353 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #46INTELLECT-31356
- ~#47Hermes 4 - Llama-3.1 405BEstimated1353
- #47gpt-oss-120b1352
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#22 of 26 (estimated)~1430 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #21Gemini 3.1 Pro1491
- ~#22Hermes 4 - Llama-3.1 405BEstimated1430
- #22Mistral Large 31412
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#108 of 19272.7%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #107Llama 3.1 Nemotron Ultra 253B v172.8%
- #108Hermes 4 - Llama-3.1 405B72.7%
- #109Ring-flash-2.072.5%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#109 of 19810.9%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #106Sarvam 105B11.0%
- #109Hermes 4 - Llama-3.1 405B10.9%
- #109Nanbeige4.1-3B10.9%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~66.0%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Hermes 4 - Llama-3.1 405BEstimated66.0%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#75 of 103 (estimated)~13.1
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #74Command A+13.1
- ~#75Hermes 4 - Llama-3.1 405BEstimated13.1
- #75Nemotron 3.5 Lightning12.9
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#76 of 11922.2%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #76Gemma 4 E2B22.2%
- #76Hermes 4 - Llama-3.1 405B22.2%
- #78Hermes 4 - Llama-3.1 70B21.6%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~55.1%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Hermes 4 - Llama-3.1 405BEstimated55.1%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~38.7%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #20Nemotron 3 Ultra44.4%
- ~#21Hermes 4 - Llama-3.1 405BEstimated38.7%
- #21Mistral Small 421.3%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#57 of 92 (estimated)~739 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #56Mistral Medium 3.5747
- ~#57Hermes 4 - Llama-3.1 405BEstimated739
- #57Ring-2.6-1T734
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#52 of 85 (estimated)~5.4%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #51Qwen3.5 122B A10B5.5%
- ~#52Hermes 4 - Llama-3.1 405BEstimated5.4%
- #52Qwen3.6 35B A3B5.2%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#64 of 91 (estimated)~9.8%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #63Nemotron 3 Super 120B A12B10.3%
- ~#64Hermes 4 - Llama-3.1 405BEstimated9.8%
- #64Claude Haiku 4.59.3%
No published score — estimated from related benchmarks