Scheduled for retirement on Mar 30, 2027. Still available at the prices below until then.
Context
200K
Max output
131K
Serving providers
17
Cheapest input
$0.02 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial AnalysisSpeed & latency
Output speed
139 t/s
tokens / second
Time to first token
0.86s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CritPt | 0% | default | — | |||||||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 1374.4 | default | — | |||||||||||||||||
| GDPval v2Frozen Aug 18, 2026 | -102.2 | default | — | |||||||||||||||||
| ||||||||||||||||||||
| GPQA Diamond | 26% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| Humanity's Last Exam | 5% | default | — | |||||||||||||||||
| IFBench | 29% | default | — | |||||||||||||||||
| LiveCodeBench | 12% | default | — | |||||||||||||||||
| Long-Context Reasoning | 18% | default | — | |||||||||||||||||
| MMLU-ProFrozen Aug 18, 2026 | 48% | default | — | |||||||||||||||||
| Omniscience | -30.9 | default | — | |||||||||||||||||
| Omniscience Accuracy | 0.1 | default | — | |||||||||||||||||
| Omniscience Non Hallucination | 0.6 | default | — | |||||||||||||||||
| SciCodeArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 13% | — | — | |||||||||||||||||
| TerminalBench Hard | 1% | default | — | |||||||||||||||||
| TerminalBench v2.1 | 1% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| τ-bench BankingArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 2% | — | — | |||||||||||||||||
| τ²-bench | 16% | default | — | |||||||||||||||||
Artificial Analysis last published scores for this benchmark on Sep 9, 2026. Scores here are frozen at that date, and models released since are not on it.
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.02 to $0.72 per 1M tokens across 17 providers, so the dearest route costs 3500% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| NebiusCheapest | $0.02 | $0.06 | 1 | |||||||||||||
Standardnebius/meta-llama/Meta-Llama-3.1-8B-Instruct
| ||||||||||||||||
| Novita | $0.02 | $0.05 | 1 | |||||||||||||
Standardnovita/meta-llama/llama-3.1-8b-instruct
| ||||||||||||||||
| Lambda AI | $0.025 | $0.04 | 1 | |||||||||||||
Standardlambda_ai/llama3.1-8b-instruct
| ||||||||||||||||
| DeepInfra | $0.03 | $0.05 | 1 | |||||||||||||
Standarddeepinfra/meta-llama/Meta-Llama-3.1-8B-Instruct
| ||||||||||||||||
| Nscale | $0.03 | $0.03 | 1 | |||||||||||||
Standardnscale/meta-llama/Llama-3.1-8B-Instruct
| ||||||||||||||||
| OpenRouter | $0.05 | $0.08 | 1 | |||||||||||||
Standardmeta-llama/llama-3.1-8b-instruct
| ||||||||||||||||
| Fireworks AI | $0.10 | $0.10 | 1 | |||||||||||||
Standardfireworks_ai/accounts/fireworks/models/llama-v3p1-8b-instruct
| ||||||||||||||||
| Friendliai | $0.10 | $0.10 | 1 | |||||||||||||
Standardfriendliai/meta-llama-3.1-8b-instruct
| ||||||||||||||||
| OCI is dearest at $0.72 / $0.72 | ||||||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Input down 95% since first tracked
Cost calculator
estimate your monthly spend on this model$9
estimated / month
Model IDs
copy the exact identifier for your platformnebius/meta-llama/Meta-Llama-3.1-8B-Instructnovita/meta-llama/llama-3.1-8b-instructlambda_ai/llama3.1-8b-instructdeepinfra/meta-llama/Meta-Llama-3.1-8B-Instructnscale/meta-llama/Llama-3.1-8B-Instructmeta-llama/llama-3.1-8b-instructfriendliai/meta-llama-3.1-8b-instructfireworks_ai/accounts/fireworks/models/llama-v3p1-8b-instructovhcloud/Llama-3.1-8B-Instructsambanova/Meta-Llama-3.1-8B-Instructhyperbolic/meta-llama/Meta-Llama-3.1-8B-Instructdatabricks/databricks-meta-llama-3-1-8b-instructperplexity/llama-3.1-8b-instructmeta.llama3-1-8b-instruct-v1:0wandb/meta-llama/Llama-3.1-8B-Instructazure_ai/Meta-Llama-3.1-8B-Instructoci/meta.llama-3.1-8b-instruct
Frequently asked questions
Llama 3.1 8B Instruct pricing, context and availabilityIs Llama 3.1 8B Instruct still available?
Yes. Llama 3.1 8B Instruct is scheduled for retirement on 2027-03-30.
How much does Llama 3.1 8B Instruct cost?
Llama 3.1 8B Instruct costs $0.02 per 1M input tokens from Nebius, and $0.03 per 1M output tokens from Nscale. Across 17 serving providers, input prices range from $0.02 to $0.72 per 1M tokens.
What is the context window of Llama 3.1 8B Instruct?
Llama 3.1 8B Instruct accepts up to 200K tokens of context and can return up to 131K output tokens.
Which providers serve Llama 3.1 8B Instruct?
Llama 3.1 8B Instruct is available from 17 serving providers, each with its own pricing and model ID. Nebius is currently the cheapest.
What can Llama 3.1 8B Instruct do?
Llama 3.1 8B Instruct supports tool use and structured output.
Other Meta models
compare pricing across the Meta lineupLlama 3.3 70B Instruct$0.12/1M · 18 providersLlama 3.1 70B Instruct$0.12/1M · 12 providersLlama 3.2 3B Instruct$0.015/1M · 11 providersLlama 4 Scout 17B 16e Instruct$0.05/1M · 11 providersLlama 3.1 405B Instruct$0.12/1M · 9 providersLlama 3 70B Instruct$0.12/1M · 9 providersLlama 3.2 1B Instruct$0.02/1M · 8 providersLlama 3.2 11B Vision Instruct$0.015/1M · 7 providers
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI