Llama 3.3 70B Instruct
Llama · text → text
Input$0.6 /M
Output$1.2 /M
← ModelsLlama 3.3 70B InstructLlama 4 Scout
Llama 3.3 70B Instruct vs Llama 4 Scout
Compare Llama 3.3 70B Instruct from Llama and Llama 4 Scout from Llama on key metrics including benchmarks, price, context length, and other model features. Access both models and hundreds of others through the AIHubMix API.
Llama 3.3 70B InstructLlama 4 Scout+ Add model · 2/4
Llama 4 Scout
Llama · text, image → text
Llama 4 Scout is a highly efficient Mixture-of-Experts (MoE) model from Meta, activating 17B out of 109B total parameters per inference. It natively supports multimodal input (text and image) and multilingual output (text and code) across 12 languages. Designed for assistant-style interaction and visual reasoning, Scout features a massive 10-million-token context window. It is instruction-tuned for tasks like multilingual chat and image understanding and is released under the Llama 4 Community License for local or commercial deployment.
Input$0.2 /M
Output$0.2 /M
Pricing & Specifications
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Llama 3.3 70B Instruct
Llama 4 Scout
Input /M
$0.6
$0.2
Output /M
$1.2
$0.2
Cache read /M
-
-
Context length
131,072
131,000
Max output
0
131,000
Time to First Token
1.9 s
0.3 s
Throughput
14.3 tok/s
2637.0 tok/s
Modalities
text
textimage
Supported Parameters
long context
toolsfunction callingstructured outputs
API Formats
Released
-
-
Promotional prices show the discounted rate; see each model page for promotion windows.
Activity Past 30 Days
Daily traffic served through AIHubMix — how demand for each model is trending.
llama-3.3-70b-instructllama-4-scout
Tokens / day
-
Requests / day
-
Performance Past 3 Days
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
llama-3.3-70b-instructllama-4-scout
Throughput (tok/s)
-
TTFT (s)
-
Uptime (%)
-
LMArena Benchmarks
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Text
llama-3.3-70b-instructllama-4-scout
1260130013401380
Overall
13171323
Coding
13461362
Math
12961309
Hard prompts
13201330
Instruction following
12921300
Multi-turn
13161320
Creative writing
12851290
Longer query
13131327
Chinese
12931315
English
13411344
Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.
Cost calculator
Estimate your monthly bill for the same workload on each model.
Llama 4 Scout
$15.00 /mo
Llama 3.3 70B Instruct
$54.00 /mo
Monthly = daily × 30. Discounted rates applied where a promotion is active.
FAQ
Which is cheaper: Llama 3.3 70B Instruct, Llama 4 Scout?
Llama 4 Scout: $0.2/M output tokens; Llama 3.3 70B Instruct: $1.2/M. Use the cost calculator above to estimate your own workload.
How do their coding arena scores compare?
Llama 4 Scout: 1362; Llama 3.3 70B Instruct: 1346 (LMArena coding leaderboard).
Which responds faster?
Llama 4 Scout: 0.3s time to first token measured on AIHubMix; see the live performance charts above for how each model behaves across the day.
How large is each context window?
Llama 3.3 70B Instruct accepts 131,072 and Llama 4 Scout accepts 131,000 input tokens.
Which one generates tokens faster?
Llama 4 Scout at 2637.0 tok/s and Llama 3.3 70B Instruct at 14.3 tok/s, measured as output throughput on AIHubMix — a separate metric from time to first token.
What inputs and capabilities does each model support?
Llama 3.3 70B Instruct accepts text input and supports long context; Llama 4 Scout accepts text and image input and supports tool calling, function calling and structured outputs.
Can I call Llama 3.3 70B Instruct and Llama 4 Scout with the same API key?
Yes. AIHubMix serves every model on this page behind one OpenAI-compatible endpoint, so switching between them is a one-line change to the model field — no second account, key or SDK.
Popular comparisons
Related model match-ups readers also look at.
AIHubMix