Llama 3.3 70B Instruct

Nov 2024
Llama 3.3 70B Instruct
Compare in
on
—per task?
$0.10 / $0.32per 1M in / out?
89 tok/sfirst token in 1.8 s?
128Kcontext

Rankingin non-thinking mode?

Strong at

B
Instructions#125 of 252
C
Tool calling#161 of 251Reasoning#347 of 360Science#264 of 336Long context#252 of 330Knowledge#241 of 332

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperGPT OSS 120BhighOpen×10-61%$0.11Memory (4-bit): 42 GB → 70 GBBetterMiniMax M2.7thinkingOpen×46.8+239%$0.11Memory (4-bit): 42 GB → 137 GBOpenGemma 4 31Bthinking×66+3%$0.10 / $0.34BetterMiMo v2.5 ProthinkingOpen×67.7+251%$0.15Memory (4-bit): 42 GB → 614 GBBetterGPT-5.6 Lunamax×71.4+190%$0.18

Run it yourself

estimate
Fits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✓ 48 GB GPU · compressed✓ 80 GB GPU · 8-bit
LicenceLlama 3.3 licencecommercial use with conditions

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam3.6#356
AA Omniscience Index-54.2#245
CritPt (physics)0#184
AA Intelligence Index 4.3.2≈7.7#267
AA Omniscience · Accuracy19#173
AA Omniscience · Hallucination Rate90.2#267
AA-LCR (long context) 1.115.7#256
GPQA Diamond49.8#268
AIME 2025 · AA archived7.7#141
Coding
Terminal-Bench Hard (version unstated)3#183
Terminal-Bench 2.14.9#138
LiveCodeBench · AA archived28.8#131
Agents & Tool Use
ITBench SRE0.6#38
τ²-bench · AA26.6#164
Writing & Language
IFBench (instructions)47.1#127

FAQ

How much does Llama 3.3 70B Instruct cost?›

$0.10 per 1M input tokens and $0.32 per 1M output tokens (cheapest host: DeepInfra).

Is Llama 3.3 70B Instruct free to use?›

There is no free API; the cheapest paid price is $0.10 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Llama 3.3 70B Instruct, and what is it best at?›

#347 of 360 in Chat & Knowledge, #179 of 243 in Coding, #161 of 251 in Agents & Tool Use, #152 of 197 in Writing & Language.

Is Llama 3.3 70B Instruct better than Nemotron 3 Super 120B A12B?›

On ITBench SRE, Llama 3.3 70B Instruct scores 0.6 vs 1.1 — about the same.

How can I use Llama 3.3 70B Instruct?›

Through the API of AkashML, Cloudflare, CoreWeave, DeepInfra, Google and 4 more.

What is the context window of Llama 3.3 70B Instruct?›

128K tokens, with up to 4K output tokens.

When was Llama 3.3 70B Instruct released? Is there a newer version?›

Released 2024-11-26. It is the current version in its line.

Is Llama 3.3 70B Instruct open source?›

Yes, the weights are open (llama3.3 licence).

Can I run Llama 3.3 70B Instruct locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can Llama 3.3 70B Instruct read images?›

No, it accepts text only.

What is Llama 3.3 70B Instruct?›

Llama 3.3 70B Instruct is an open-weights language model from Meta, released Nov 2024.

What are the best alternatives to Llama 3.3 70B Instruct?›

GPT OSS 120B (cheaper), MiniMax M2.7 (better), Gemma 4 31B (open weights).