Meta: Llama 4 Scout

Text input Image input Text output
Author's Description

Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...

Key Specifications
Cost
$$
Context
327K
Parameters
17B
Released
Apr 05, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Seed Min P Response Format Temperature Presence Penalty Top Logprobs Tools Frequency Penalty Top P Logprobs Stop Tool Choice Max Tokens Logit Bias
Features

This model supports the following features:

Response Format Tools
Performance Summary

Meta's Llama 4 Scout 17B Instruct (16E) demonstrates a balanced performance profile, excelling in cost-effectiveness and reliability while showing mixed results in accuracy across various tasks. The model performs among the fastest models, ranking in the 63rd percentile for speed, and offers competitive pricing, placing in the 77th percentile. Its reliability is notably strong, with a 93% success rate, indicating consistent and usable responses. Llama 4 Scout shows exceptional performance in Email Classification (99% accuracy, 91st percentile), highlighting its strength in categorization tasks. It also performs well in General Knowledge (97% accuracy) and Ethics (98% accuracy), suggesting a solid understanding of factual information and ethical principles. However, the model exhibits significant weaknesses in Mathematics (39% accuracy, 15th percentile) and Instruction Following (38.7% accuracy, 31st percentile), indicating challenges with complex logical operations and multi-step directives. Its hallucination rate is also a concern at 68% accuracy, suggesting a tendency to generate information rather than acknowledge uncertainty. While its Coding performance is moderate (79.5% accuracy), its Reasoning capabilities are average (58% accuracy).

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.08
Completion $0.3

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Lambda
Lambda | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
DeepInfra
DeepInfra | meta-llama/llama-4-scout-17b-16e-instruct 327K $0.08 / 1M tokens $0.3 / 1M tokens
Kluster
Kluster | meta-llama/llama-4-scout-17b-16e-instruct 131K $0.08 / 1M tokens $0.3 / 1M tokens
GMICloud
GMICloud | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Parasail
Parasail | meta-llama/llama-4-scout-17b-16e-instruct 158K $0.08 / 1M tokens $0.3 / 1M tokens
Cent-ML
Cent-ML | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Novita
Novita | meta-llama/llama-4-scout-17b-16e-instruct 131K $0.08 / 1M tokens $0.3 / 1M tokens
Groq
Groq | meta-llama/llama-4-scout-17b-16e-instruct 131K $0.11 / 1M tokens $0.34 / 1M tokens
BaseTen
BaseTen | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Fireworks
Fireworks | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Together
Together | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Google
Google | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.25 / 1M tokens $0.7 / 1M tokens
SambaNova
SambaNova | meta-llama/llama-4-scout-17b-16e-instruct 8K $0.08 / 1M tokens $0.3 / 1M tokens
Cerebras
Cerebras | meta-llama/llama-4-scout-17b-16e-instruct 32K $0.08 / 1M tokens $0.3 / 1M tokens
BaseTen
BaseTen | meta-llama/llama-4-scout-17b-16e-instruct 1M $0.08 / 1M tokens $0.3 / 1M tokens
Friendli
Friendli | meta-llama/llama-4-scout-17b-16e-instruct 447K $0.08 / 1M tokens $0.3 / 1M tokens
DeepInfra
DeepInfra | meta-llama/llama-4-scout-17b-16e-instruct 327K $0.08 / 1M tokens $0.3 / 1M tokens
Novita
Novita | meta-llama/llama-4-scout-17b-16e-instruct 131K $0.18 / 1M tokens $0.59 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by meta-llama