SI Index Data as of Oct 9, 2026

Llama 4 Maverick 17B Instruct: Price, Context, Benchmarks, and Release Details

Meta provisional listing open weights · first seen Oct 8, 2026 · score computed Oct 9, 2026, 01:50 UTC
30.0
SI Score (method si-v3-retained-evidence-2)
#139 of 140 ranked
78% confidence 78 percent, Medium confidence — 6 of 9 expected sources in
Coverage 63% of expected source weight · 100% confidence at 80% coverage

Pillar breakdown

Coding (weight 40 percent) 10.4
Math (weight 15 percent) 55.5
Preference (weight 15 percent) —
Reasoning (weight 30 percent) 9.2

Pillar weights: reasoning 30% · math 15% · coding 40% · preference 15%. Benchmark results use fixed 0–100 scales before averaging; incomplete evidence is shrunk toward 50.

Facts

Input price / 1M
not yet reported
Output price / 1M
not yet reported
Context window
1Mmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Max output
16.4Kmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Released
Apr 5, 2025models.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Open weights
Yesmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
License
not yet reported
Input modalities
text, imagemodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗

Benchmark results

11 results
Benchmark Raw result Normalized (0–100) Pillar
aider polyglot 15.6%Aider polyglotPublished source fact [variant] Aider polyglot; 225 cases; 2 attemptsPublished Apr 6, 2025 Retrieved Oct 9, 2026 · Apache-2.0
Open source ↗ 		15.6 coding
arc agi v1 public eval 7.1%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] Llama-4-Maverick-17B-128E-Instruct-FP8-togetherPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		7.1 reasoning
arc agi v1 semi private 4.4%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] Llama-4-Maverick-17B-128E-Instruct-FP8-togetherPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		4.4 reasoning
arc agi v2 public eval 0%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] Llama-4-Maverick-17B-128E-Instruct-FP8-togetherPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		0.0 reasoning
arc agi v2 semi private 0%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] Llama-4-Maverick-17B-128E-Instruct-FP8-togetherPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		0.0 reasoning
gpqa diamond 67.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Apr 8, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		67.0 reasoning
hle scale 5.7%Humanity’s Last ExamPotential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. [variant] Published Apr 10, 2025 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		5.7 reasoning
math level 5 73.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Apr 8, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		73.0 math
otis mock aime 2024 2025 20.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Apr 8, 2025 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		20.6 math
swe bench pro public 5.2%Official model cards via models.devLab-reported; metric resolve rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] public Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ 		5.2 coding
swe bench pro public 5.2%SWE-bench Pro (public)Published steward score [variant] Published Jan 27, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 		5.2 coding

Normalization uses fixed absolute 0–100 scales for each unit, independently of other models that have a result on each benchmark. Coverage and evidence breadth still affect the composite; compare the evaluation conditions before reading a small score gap as decisive. “lab-reported” marks the provider's own published figure.

Sources: in and pending

in Reported (6)

  • Aider polyglot · arrived Oct 8, 2026
  • ARC Prize · arrived Oct 8, 2026
  • Epoch AI Benchmarking · arrived Oct 8, 2026
  • Humanity’s Last Exam · arrived Oct 8, 2026
  • Official model cards via models.dev · arrived Oct 8, 2026
  • SWE-bench Pro (public) · arrived Oct 8, 2026

pending Awaiting (3)

  • LiveBench · carries 11% of expected weight
  • LMArena / Arena · carries 21% of expected weight
  • Terminal-Bench · carries 5% of expected weight

The confidence % rises as pending sources publish. Some sources never cover some models — that is why 100% confidence arrives at 80% of expected weight, not at full coverage.