Llama 3.1 70B Instruct
Meta · released Jul 16, 2024 · meta-llama/Meta-Llama-3.1-70B-Instruct
- Type
- Open weightsLlama 3.1 Community License
- Params
- 70.6B
- Context
- 131K
about 98K words of context · download allowed, licence restricts use
Our take
Written Sep 2, 2026Llama 3.1 is a 70.6-billion-parameter text model from Meta released in 2024 with a 131,072-token request limit. It is a dense, full-parameter workhorse with strong instruction-following scores and several hosts offering it at the low end of their price range.
Pick this for long-context text work at 131,072 tokens, or budget-conscious production use where the cheapest tracked hosts charge the same for input and output. Use it when instruction-following accuracy matters more than graduate-level science reasoning. Skip it if you need image or audio input, if the licence terms pose legal risk, or if you want sparse routing to cut per-token compute.
The case for it
- IFEval instruction-following accuracy of 86.7% — its highest measured score.
- Arena coding score 40 points above its own general text rating, making code its relative strength.
- 131,072-token request limit enables single-prompt analysis of book-length documents.
- Several hosts price it at the bottom of their range, well under half the cost of the most expensive tracked offer.
The case against it
- GPQA Diamond graduate-level science at 14.2% correct — its lowest score, far below its other marks.
- Most Arena categories sit below its own average; only coding exceeds it.
- All 70.6 billion parameters are active per token, with no sparse routing to reduce compute.
How good is it?
An open-weight text model for chat and general assistance, though it trails most models on everyday questions, writing and coding.
- getting answers to everyday questionsArena Text (overall) · 149th of 168
- drafts, rewrites and editingArena Creative Writing · 148th of 168
- writing and completing codeArena Coding · 149th of 168
EverydayGeneral questions and everyday reasoning
Arena Text (overall)149th of 168 · 1293
CodingWriting and fixing code on its own
Arena Coding149th of 168 · 1333
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing148th of 168 · 1257
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Max (32-core GPU) · 64 GB
Borderline fit on an estimated size. It leaves 0.6 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked between 4 hours and 10 hours ago — each listing carries its own date.
- per 1M tokens
- $0.40 in / $0.40 out
- Context served
- 131K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $0.40 / $0.40checked 4 hours ago | 131K | not measured | Unknown | Unknown | Unknown |
| DeepInfraturbo tierfp8Through OpenRouter | $0.40 / $0.40checked 10 hours ago | 131K16K max reply | 17 tok/s | No | No | Unknown |
| Amazon BedrockThrough OpenRouter | $0.72 / $0.72checked 4 hours ago | 131K8K max reply | 25 tok/s | No | No | Confirmed |
DeepInfra
Amazon BedrockAcross the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 does not say. 1 appears in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| DeepInfraturbo · fp8Through OpenRouter | ✓ | ✓ | ✓ |
| Amazon BedrockThrough OpenRouter | ✗ | ✗ | ✗ |
Tool calling: 2 of 3 listings say yes, 1 says no. JSON output: 2 of 3 listings say yes, 1 says no. Strict schema: 2 of 3 listings say yes, 1 says no.
Models people weigh against Llama 3.1 70B Instruct
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 listings does not say whether it trains on prompts.
- We hold no cached-input rate for any of its listings.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsLlama 3.1 Community License, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Llama 3.1 Community License
Commercial use below 700M MAU. Notably allows using outputs to improve other models, which earlier Llama licenses banned. Derivative names must start with "Llama".
Identifiers
- Hugging Face
- meta-llama/Meta-Llama-3.1-70B-Instruct
- Takes in, gives back
- Text in, text out
- Catalogue slug
- meta-llama-llama-3-1-70b-instruct