Phi-4
- Input
- $0.125/M tok
- Output
- $0.50/M tok
- Blended
- $0.219/M tok · 3:1
- Speed
- 40tok/s
- Context
- 16Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Phi-4. 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
- 1× B200 180GB
- BF16 weights
- Throughput
- ~7,374 tok/s
- many requests batched
- Price
- $0.12–1.08 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, ~170 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- BF16 weights (14.7B params) + 25% KV-cache headroom ≈ 37 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~7,374 output tok/s aggregate at batch 211 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~36,161 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
#132 of 13923.5%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #131Granite 4.0 Micro24.8%
- #132Phi-423.5%
- #133Apertus 8B Instruct22.4%
artificialanalysis.ai · 2026-09-24
MMMLUEstimated
~last of 5 (estimated)below measured range (<81.3%)
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Verified
#55 of 561233 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #54NVIDIA Nemotron 3 Nano 30B A3B1286
- #55Phi-41233
- #56Olmo 3.1 32B Think1231
lmarena.ai · 2026-09-13
EQ-Bench Creative Writing v3 (Elo)Estimated
~#24 of 26 (estimated)~1149 Elo
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#140 of 19257.5%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #139Llama 4 Scout58.7%
- #140Phi-457.5%
- #141Ministral 3 14B57.2%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~40.5%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Phi-4Estimated40.5%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#102 of 103 (estimated)~4.6
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #100NVIDIA Nemotron 3 Nano 4B4.8
- ~#102Phi-4Estimated4.6
- #102K2 Horizon 0.9B3.0
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#110 of 1190.0%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #110Olmo 3.1 32B Think0.0%
- #110Phi-40.0%
- #110Reka Flash 30.0%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#13 of 15 (estimated)~51.7%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #12Nemotron 3.5 Super VL NVFP452.3%
- ~#13Phi-4Estimated51.7%
- #13Nemotron 3.5 Super VL51.1%
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)~35.4%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #20Nemotron 3 Ultra44.4%
- ~#21Phi-4Estimated35.4%
- #21Mistral Small 421.3%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#63 of 92 (estimated)~603 Elo
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#85 of 85 (estimated)~0.2%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~last of 91 (estimated)below measured range (<3.3%)
No published score — estimated from related benchmarks