Llama 4 Maverick
- Input
- $0.26/M tok
- Output
- $0.91/M tok
- Blended
- $0.422/M tok · 3:1
- Speed
- 38tok/s
- Context
- 1Mtokens
- Cost per task
- ~$0.38/taskest.
Previous-generation open-weights model, for context. Price is the median third-party host price.
Rankings
Coding
Verified#124 / 137rank by quality21.9±6.5quality / 1003/7 benchmarks measuredWriting & chat
Verified#125 / 162rank by quality23.5±3.3quality / 1004/5 benchmarks measuredVision
Mixed#76 / 98rank by quality31.1±6.8quality / 1002/6 benchmarks measuredHard reasoning
Verified#168 / 218rank by quality21.1±3.3quality / 1004/4 benchmarks measuredAgents
Verified#177 / 187rank by quality12.3±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
- 8× H200 141GB
- FP8 weights
- Throughput
- ~25,249 tok/s
- many requests batched
- Price
- $0.14–1.17 /M tok
- busy → light use
- On your own machine
- Mac Studio M5 Ultra 512GB (int4, ~99 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (401.6B params, 17B active per token) + 40% KV-cache headroom ≈ 562 GB
- 8× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
- ~25,249 output tok/s aggregate at batch 908 (bandwidth-bound); MoE compute scales with active params
- Blended 3:1 input:output; prefill ~220,018 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
~#17 of 21 (estimated)~64.8%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #16Nemotron 3 Ultra70.7%
- ~#17Llama 4 MaverickEstimated64.8%
- #17Intern-S2-Preview (35B-A3B)64.0%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~44.4%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23Llama 4 MaverickEstimated44.4%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#43 of 46 (estimated)~35.3%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #42Nex-N2.5-Mini36.1%
- ~#43Llama 4 MaverickEstimated35.3%
- #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%
- #63Ling-3.0-flash-VL0.0%
- #63Llama 4 Maverick0.0%
- #63Llama 4 Scout0.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBenchVerified
#22 of 2739.7%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #21Devstral 244.8%
- #22Llama 4 Maverick39.7%
- #23Ministral 3 14B35.1%
artificialanalysis.ai · 2026-09-24
SciCodeVerified
#102 of 11831.7%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #101Nemotron 3.5 Lightning32.1%
- #102Llama 4 Maverick31.7%
- #103Granite 4.2 8B31.5%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~27.2
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25Llama 4 MaverickEstimated27.2
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Writing & chat
IFBenchVerified
#85 of 13943.0%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #84LongCat Flash Lite43.1%
- #85Llama 4 Maverick43.0%
- #86LFM2.5-1.2B-Thinking41.8%
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)Verified
#48 of 561311 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #47INTELLECT-3Few votes1327
- #48Llama 4 Maverick1311
- #49Llama 3.1 Nemotron Ultra 253B v1Few votes1306
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#50 of 561327 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #49Olmo 3.1 32B Instruct1330
- #50Llama 4 Maverick1327
- #51Molmo2-8BFew votes1322
lmarena.ai · 2026-09-24
#25 of 25860 Elo
eqbench.com · 2026-09-24
Vision
MMMU-ProVerified
#59 of 8762.1%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #58Command A+63.2%
- #59Llama 4 Maverick62.1%
- #59Qwen3.5 4B62.1%
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%
- ~#12Llama 4 MaverickEstimated<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)Estimated
~#41 of 48 (estimated)~2.4%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #40Mistral Medium 3.52.8%
- ~#41Llama 4 MaverickEstimated2.4%
- #41Ministral 3 14B1.2%
No published score — estimated from related benchmarks
ChartographyEstimated
~last of 25 (estimated)below measured range (<9.0%)
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#121 of 19267.1%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #120Claude Haiku 4.567.2%
- #121Llama 4 Maverick67.1%
- #122DiffusionGemma 26B A4B66.9%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#169 of 1984.9%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #162Qwen3.5 2B5.0%
- #169Llama 4 Maverick4.9%
- #169NVIDIA Nemotron 3 Nano 4B4.9%
artificialanalysis.ai · 2026-09-24
Agents
τ²-bench (Telecom)Verified
#86 of 11917.8%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #84LFM2.5-1.2B-Thinking19.6%
- #86Llama 4 Maverick17.8%
- #87Granite 4.0 H Small17.3%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~55.3%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Llama 4 MaverickEstimated55.3%
- #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)~46.7%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #19Mistral Medium 3.548.6%
- ~#20Llama 4 MaverickEstimated46.7%
- #20Nemotron 3 Ultra44.4%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#80 of 92 (estimated)~350 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #79Nova 2.0 Lite354
- ~#80Llama 4 MaverickEstimated350
- #80Mistral Small 4349
No published score — estimated from related benchmarks
AutomationBench-AAVerified
#79 of 840.2%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #79K2 Horizon 0.9B0.2%
- #79Llama 4 Maverick0.2%
- #79Llama 4 Scout0.2%
artificialanalysis.ai · 2026-09-24