Magistral Small 1.2
- Input
- $0.50/M tok
- Output
- $1.50/M tok
- Blended
- $0.75/M tok · 3:1
- Speed
- 75tok/s
- Context
- 128Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Magistral Small 1.2. Price = AA median host price.
Rankings
Writing & chat
Verified#95 / 162rank by quality36.0±9.5quality / 1001/5 benchmarks measuredVision
Verified#75 / 98rank by quality31.5±8.2quality / 1001/6 benchmarks measuredHard reasoning
Verified#131 / 218rank by quality35.1±4.2quality / 1003/4 benchmarks measuredAgents
Verified#149 / 187rank by quality15.4±5.9quality / 1003/7 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
- ~4,962 tok/s
- many requests batched
- Price
- $0.19–1.67 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, ~100 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- BF16 weights (24B params) + 25% KV-cache headroom ≈ 60 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~4,962 output tok/s aggregate at batch 142 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~22,148 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
#81 of 13944.4%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #80Ring-2.6-1T44.6%
- #81Magistral Small 1.244.4%
- #82LFM2.5-1.2B-Instruct43.8%
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
~#44 of 57 (estimated)~1366 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #43Qwen3 Next 80B A3B Instruct1386
- ~#44Magistral Small 1.2Estimated1366
- #44GLM-4.6VFew votes1361
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#44 of 57 (estimated)~1385 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #43Qwen3 Next 80B A3B Instruct1399
- ~#44Magistral Small 1.2Estimated1385
- #44GLM-4.6VFew votes1378
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#23 of 26 (estimated)~1403 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #22Mistral Large 31412
- ~#23Magistral Small 1.2Estimated1403
- #23Gemma 4 31B1368
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#67 of 8755.5%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #66Mistral Large 355.7%
- #67Magistral Small 1.255.5%
- #67Qwen3 Omni 30B A3B Instruct55.5%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~last of 12 (estimated)below measured range (<74.5%)
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #11Gemini 3.5 Flash-Lite74.5%
- ~#12Magistral Small 1.2Estimated<74.5%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#6 of 8 (estimated)~77.0%
- #1MiniMax-M391.6%
- #2Kimi K391.1%
- #5Claude Haiku 4.579.6%
- ~#6Magistral Small 1.2Estimated77.0%
- #6Mistral Small 476.4%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#38 of 48 (estimated)~3.9%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #37Qwen3.6 35B A3B6.6%
- ~#38Magistral Small 1.2Estimated3.9%
- #38Claude Haiku 4.53.8%
No published score — estimated from related benchmarks
ChartographyEstimated
~last of 25 (estimated)below measured range (<9.0%)
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #24Mistral Large 39.0%
- ~#25Magistral Small 1.2Estimated<9.0%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#123 of 19266.3%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #122DiffusionGemma 26B A4B66.9%
- #123Magistral Small 1.266.3%
- #124Falcon-H1R-7B66.1%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#139 of 1986.4%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #139Granite 4.0 H 350M6.4%
- #139Magistral Small 1.26.4%
- #141Gemma 4 12B6.3%
artificialanalysis.ai · 2026-09-24
AIME (latest)Verified
#21 of 3380.3%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #20Claude Haiku 4.583.7%
- #21Magistral Small 1.280.3%
- #22Phi-4-reasoning-plus78.0%
artificialanalysis.ai · 2026-09-24
AA Intelligence IndexEstimated
~#86 of 103 (estimated)~10.2
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #85Trinity Large Thinking10.8
- ~#86Magistral Small 1.2Estimated10.2
- #86North Mini Code9.9
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#62 of 11927.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #62K2-V227.8%
- #62Magistral Small 1.227.8%
- #65Ministral 3 14B27.2%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~56.9%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Magistral Small 1.2Estimated56.9%
- #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)~48.5%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #19Mistral Medium 3.548.6%
- ~#20Magistral Small 1.2Estimated48.5%
- #20Nemotron 3 Ultra44.4%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#41 of 92 (estimated)~1019 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #40Kimi K2.7 Code1025
- ~#41Magistral Small 1.2Estimated1019
- #41Nemotron 3 Ultra1013
No published score — estimated from related benchmarks