LIVE
Introducing Cosmos 3 Edge20/07/26 · Hugging Face|Kimi K3, Qwen 3.8, and Anthropic's (Potential) Unravelling20/07/26 · Anthropic|At SIGGRAPH, NVIDIA Advances Graphics and Simulation With Agentic and Physical AI20/07/26 · NVIDIA|China's open-weights AI strategy is winning20/07/26|Bristol Myers Squibb Building Life Science Industry’s Most Advanced AI Factory on NVIDIA Vera Rubin20/07/26 · NVIDIA|Safety and alignment in an era of long-horizon models20/07/26 · OpenAI|Claude Fable produced a counterexample to the Jacobian Conjecture20/07/26 · Anthropic|Apply for Anthropic’s AI for Science rare disease research grants20/07/26 · Anthropic|AI advice made people less accurate but more confident – sudy19/07/26|Vincentwei1021/video-shotcraft: AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 106 shot recipe cards, 161 motion previews, a production-ready template19/07/26 · Anthropic|Claude Code uses Bun written in Rust now19/07/26 · Anthropic|OpenAI reduces Codex Model Context Size from 372k to 272k19/07/26 · OpenAI|Introducing Cosmos 3 Edge20/07/26 · Hugging Face|Kimi K3, Qwen 3.8, and Anthropic's (Potential) Unravelling20/07/26 · Anthropic|At SIGGRAPH, NVIDIA Advances Graphics and Simulation With Agentic and Physical AI20/07/26 · NVIDIA|China's open-weights AI strategy is winning20/07/26|Bristol Myers Squibb Building Life Science Industry’s Most Advanced AI Factory on NVIDIA Vera Rubin20/07/26 · NVIDIA|Safety and alignment in an era of long-horizon models20/07/26 · OpenAI|Claude Fable produced a counterexample to the Jacobian Conjecture20/07/26 · Anthropic|Apply for Anthropic’s AI for Science rare disease research grants20/07/26 · Anthropic|AI advice made people less accurate but more confident – sudy19/07/26|Vincentwei1021/video-shotcraft: AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 106 shot recipe cards, 161 motion previews, a production-ready template19/07/26 · Anthropic|Claude Code uses Bun written in Rust now19/07/26 · Anthropic|OpenAI reduces Codex Model Context Size from 372k to 272k19/07/26 · OpenAI|
🇺🇸 United States🧠 GeneralistReleased April 2025

Llama 4 Behemoth

Meta
Open-sourceFreeSelf-host
Verdict

A solid pick for self-hosting and full data control.

Overview

Llama 4 Behemoth is Meta's largest model, natively multimodal with 1M context. The Llama Community license enables commercial use for organizations under 700M MAU. Ideal for enterprise-grade self-host setups. Much slower than closed frontiers.

Skill profile

FrenchReasoningSpeedCreativitySafety
Arena Elo1330
Verified benchmarks
GPQA66%
SWE-Bench45%
MATH86%
MMLU87%
HumanEval88.5%

What do these scores mean?

GPQA66%very good

PhD-level science questions (physics, chemistry, biology), with no tool access.

SWE-Bench45%fair

Resolving real GitHub issues under real conditions (SWE-bench Verified).

MATH86%excellent

Competition-level math problems.

MMLU87%excellent

General knowledge across dozens of academic subjects.

HumanEval88.5%excellent

Python code generation from specifications.

Strengths

  • Excellent at code
  • Strong French quality
  • Very long context
  • Has a free tier

Limitations

  • Average speed
  • No native GDPR guarantee

Who is it for

A good fit if…
  • you want to control cost or self-host
  • you work in French
Skip it if…
  • you handle sensitive EU data
  • you need very fast responses

Ideal use cases

  • Large-scale pro self-host
  • Academic research
  • Multimodal long context

Access & availability

Paid API per tokenFree tier availableSelf-hostable (open weights)

Key specifications

Context
1M
Input price
$3.00 $/M
Output price
$9.00 $/M
Speed
40 tok/s
Price not auditedBenchmarks not audited

Estimate your monthly cost

Per-token API pricing
$154/ month with Llama 4 Behemoth

For the same usage

Indicative estimate based on standard API rates (excluding caching, batch and volume discounts). Always check the official pricing before committing.

Privacy

Not GDPR-compliantHosting : US/selfAnonymizable data
Advanced data · for experts
Architecture, modalities, detailed cost, full benchmarks
Architecture
MoE
Size
~2T (288B actifs MoE)
Cutoff date
Nov 2024
Inputs
text, image
Outputs
text, code
License
llama-community
Hosting
US/self
Hallucination score
3/5
Estimated cost (API)
Typical exchange (~3k in / 1k out)
≈ $0.0180
1M in + 1M out
≈ $12.00
All benchmarks
Arena Elo1330
MMLU87.0%
GPQA66.0%
HumanEval88.5%
SWE-Bench45.0%
MATH86.0%

Similar models