Llama 3.1 Nemotron Ultra 253B v1
- Input
- —
- Output
- —
- Speed
- ~110tok/sest.
- Context
- 128Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Llama 3.1 Nemotron Ultra 253B v1 (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. No API sells this model, so this is its price on the chart.
- Hardware
- 8× H100 80GB
- FP8 weights
- Throughput
- ~4,224 tok/s
- many requests batched
- Price
- $0.88–8.40 /M tok
- busy → light use
- On your own machine
- Mac Studio M5 Ultra 512GB (int4, ~7 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (253B params) + 25% KV-cache headroom ≈ 316 GB
- 8× H100 80GB at $2.69–$6.88/GPU-hour on-demand (2026-09-24)
- ~4,224 output tok/s aggregate at batch 123 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~14,784 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
#93 of 13938.2%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #92ERNIE 4.5 300B A47B39.1%
- #93Llama 3.1 Nemotron Ultra 253B v138.2%
- #94Devstral 238.1%
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%
- ~#5Llama 3.1 Nemotron Ultra 253B v1Estimated<81.3%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Few votesVerified
#49 of 561306 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #48Llama 4 Maverick1311
- #49Llama 3.1 Nemotron Ultra 253B v1Few votes1306
- #50Molmo2-8BFew votes1305
lmarena.ai · 2026-09-13
LMArena Text (Elo)Few votesVerified
#48 of 561348 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #47gpt-oss-120b1352
- #48Llama 3.1 Nemotron Ultra 253B v1Few votes1348
- #49Olmo 3.1 32B Instruct1330
lmarena.ai · 2026-09-13
EQ-Bench Creative Writing v3 (Elo)Estimated
~#24 of 26 (estimated)~1255 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #23Gemma 4 31B1368
- ~#24Llama 3.1 Nemotron Ultra 253B v1Estimated1255
- #24gpt-oss-120b961
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#107 of 19272.8%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #106Quasar 438B73.2%
- #107Llama 3.1 Nemotron Ultra 253B v172.8%
- #108Hermes 4 - Llama-3.1 405B72.7%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#131 of 1987.4%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #131LFM2.5-24B-A2B7.4%
- #131Llama 3.1 Nemotron Ultra 253B v17.4%
- #133Nova 2.0 Omni7.0%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~61.9%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Llama 3.1 Nemotron Ultra 253B v1Estimated61.9%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#86 of 103 (estimated)~10.6
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #85Trinity Large Thinking10.8
- ~#86Llama 3.1 Nemotron Ultra 253B v1Estimated10.6
- #86North Mini Code9.9
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#103 of 11911.4%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #103Apertus 8B Instruct11.4%
- #103Llama 3.1 Nemotron Ultra 253B v111.4%
- #105LFM2 24B A2B11.1%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~54.4%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Llama 3.1 Nemotron Ultra 253B v1Estimated54.4%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
- #1GLM-5.3 Flash59.1%
- #2Kimi K358.3%
- ~#3Llama 3.1 Nemotron Ultra 253B v1Estimated53.8%
- #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~38.6%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #20Nemotron 3 Ultra44.4%
- ~#21Llama 3.1 Nemotron Ultra 253B v1Estimated38.6%
- #21Mistral Small 421.3%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#61 of 92 (estimated)~663 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #60MiniCPM5-2B694
- ~#61Llama 3.1 Nemotron Ultra 253B v1Estimated663
- #61Nemotron 3.5 Lightning624
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#63 of 85 (estimated)~1.9%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #62Granite 4.2 8B2.1%
- ~#63Llama 3.1 Nemotron Ultra 253B v1Estimated1.9%
- #63Command A+1.8%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#69 of 91 (estimated)~7.6%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #68Granite 4.2 8B7.6%
- ~#69Llama 3.1 Nemotron Ultra 253B v1Estimated7.6%
- #69LFM2.5-2.6B7.2%
No published score — estimated from related benchmarks