Ministral 3 8B
- Input
- $0.15/M tok
- Output
- $0.15/M tok
- Blended
- $0.15/M tok · 3:1
- Speed
- 87tok/s
- Context
- 256Ktokens
- Cost per task
- ~$0.06/taskest.
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Ministral 3 8B. Price = AA median host price.
Rankings
Coding
Verified#134 / 137rank by quality15.1±6.5quality / 1003/7 benchmarks measuredWriting & chat
Verified#139 / 162rank by quality18.7±9.5quality / 1001/5 benchmarks measuredVision
Verified#85 / 98rank by quality20.7±6.3quality / 1002/6 benchmarks measuredHard reasoning
Verified#185 / 218rank by quality15.7±3.3quality / 1004/4 benchmarks measuredAgents
Verified#158 / 187rank by quality14.2±4.3quality / 1004/7 benchmarks measured
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
- ~10,124 tok/s
- many requests batched
- Price
- $0.083–0.74 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, ~280 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- BF16 weights (8.92B params) + 25% KV-cache headroom ≈ 22 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~10,124 output tok/s aggregate at batch 256 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~59,592 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
SWE-bench VerifiedEstimated
~#20 of 21 (estimated)~59.4%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #19NVIDIA Nemotron-Labs Dory1.0 4B59.5%
- ~#20Ministral 3 8BEstimated59.4%
- #20LFM2.5-24B-A2B51.0%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~44.6%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23Ministral 3 8BEstimated44.6%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#43 of 46 (estimated)~31.3%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #42Nex-N2.5-Mini36.1%
- ~#43Ministral 3 8BEstimated31.3%
- #43Hy328.0%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Verified
#63 of 1010.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #63Ministral 3 3B0.0%
- #63Ministral 3 8B0.0%
- #63Mistral Large 30.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchVerified
#25 of 2730.3%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #24Devstral Small 234.8%
- #25Ministral 3 8B30.3%
- #26Llama 4 Scout29.9%
artificialanalysis.ai · 2026-09-24
SciCodeVerified
#114 of 11820.7%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #113Llama 4 Scout21.3%
- #114Ministral 3 8B20.7%
- #115Qwen3.8 Flash-Next NVFP418.8%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~27.3
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25Ministral 3 8BEstimated27.3
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Writing & chat
IFBenchVerified
#118 of 13929.1%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #117Nova Micro29.4%
- #118Ministral 3 8B29.1%
- #118Qwen3.5 2B29.1%
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)Estimated
~#49 of 57 (estimated)~1308 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #48Llama 4 Maverick1311
- ~#49Ministral 3 8BEstimated1308
- #49Llama 3.1 Nemotron Ultra 253B v1Few votes1306
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#49 of 57 (estimated)~1332 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #48Llama 3.1 Nemotron Ultra 253B v1Few votes1348
- ~#49Ministral 3 8BEstimated1332
- #49Olmo 3.1 32B Instruct1330
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#24 of 26 (estimated)~1247 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #23Gemma 4 31B1368
- ~#24Ministral 3 8BEstimated1247
- #24gpt-oss-120b961
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#74 of 8746.0%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #73Ministral 3 14B49.8%
- #74Ministral 3 8B46.0%
- #75Devstral Small 244.6%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~last of 12 (estimated)below measured range (<74.5%)
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~last of 8 (estimated)below measured range (<75.8%)
No published score — estimated from related benchmarks
GDP.pdf (AA)Verified
#45 of 470.8%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #43Devstral Small 21.0%
- #45Ministral 3 8B0.8%
- #46Ministral 3 3B0.0%
artificialanalysis.ai · 2026-09-24
ChartographyEstimated
~last of 25 (estimated)below measured range (<9.0%)
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#159 of 19247.1%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #158LFM2 24B A2B47.4%
- #159Ministral 3 8B47.1%
- #160Nemotron 3 Nano Omni 30B A3B Reasoning46.9%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#181 of 1984.3%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #179Reka Flash 34.4%
- #181Ministral 3 8B4.3%
- #181Molmo2-8B4.3%
artificialanalysis.ai · 2026-09-24
Agents
τ²-bench (Telecom)Verified
#66 of 11926.6%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #66Llama 3.3 Instruct 70B26.6%
- #66Ministral 3 8B26.6%
- #69Gemma 4 E4B26.0%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~56.0%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Ministral 3 8BEstimated56.0%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#20 of 22 (estimated)~48.0%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #19Mistral Medium 3.548.6%
- ~#20Ministral 3 8BEstimated48.0%
- #20Nemotron 3 Ultra44.4%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#88 of 91200 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #87Nemotron 3 Nano Omni 30B A3B Reasoning216
- #88Ministral 3 8B200
- #89Granite 4.2 3B188
artificialanalysis.ai · 2026-09-24