Phi-4-mini-flash-reasoning
- Input
- —
- Output
- —
- Speed
- ~56tok/sest.
- Context
- 66Ktokens
3.8B hybrid (SambaY) math-reasoning model. No API price found.
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× B200 180GB
- BF16 weights
- Throughput
- ~16,098 tok/s
- many requests batched
- Price
- $0.046–0.41 /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)
- BF16 weights (3.85B params) + 25% KV-cache headroom ≈ 10 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~16,098 output tok/s aggregate at batch 256 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~138,068 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.
Hard reasoning
GPQA DiamondLab-reported
#161 of 19245.1%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #160Nemotron 3 Nano Omni 30B A3B Reasoning46.9%
- #161Phi-4-mini-flash-reasoning45.1%
- #162Qwen3.5 2B43.8%
huggingface.co · 2025-07-09
Humanity's Last ExamEstimated
~#173 of 199 (estimated)~4.8%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #171Nemotron 3 Nano Omni 30B A3B Reasoning4.8%
- ~#173Phi-4-mini-flash-reasoningEstimated4.8%
- #173Gemma 4 E2B4.7%
No published score — estimated from related benchmarks
AIME (latest)Lab-reported
#26 of 3333.6%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #25Devstral Small 234.3%
- #26Phi-4-mini-flash-reasoning33.6%
- #27Ministral 3 8B31.7%
huggingface.co · 2025-07-09
AA Intelligence IndexEstimated
~#103 of 103 (estimated)~2.9
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #102K2 Horizon 0.9B3.0
- ~#103Phi-4-mini-flash-reasoningEstimated2.9
No published score — estimated from related benchmarks