Llama 3 Instruct 8B
Llama 3 Instruct 8B has too few ranked results yet to rate it on any capability.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability.
No capability has enough ranked results to rate yet.
Price
$0.04input$0.14outputper million tokens
From Artificial Analysis · All prices
Evidence
73results on59benchmarks
- 21 independently verified
- 14 aggregator
- 38 cross-referenced
From 17 sources · latest Oct 8, 2026 · How verification works
API features
Structured outputs
As listed by OpenRouter
Llama 3 Instruct 8B benchmark results
73 results on 59 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
0 of 6 ranked benchmarks measured
- 0.00Oct 8, 2026
- GPQA Diamond29.60Oct 8, 2026gpqa
- Humanity's Last Exam5.06Oct 8, 2026aa_hle
Show 2 more reasoning resultsHide 2 reasoning results
- GPQA Diamond32.40Aug 29, 2026GPQA
- GPQA Diamond25.80Aug 26, 2026GPQA
0 of 10 ranked benchmarks measured
- LiveBench · Coding20.00Aug 23, 2026livebench_coding@2025-04-07
- Terminal-Bench Hard0.00Oct 8, 2026aa_terminalbench_hard
- SciCode11.94Sep 4, 2026aa_scicode
0 of 3 ranked benchmarks measured
- 0.00Oct 8, 2026
0 of 3 ranked benchmarks measured
- LiveBench · Instruction Following65.67Aug 23, 2026livebench_instruction_following@2025-04-07
- IFBench24.56Oct 8, 2026aa_ifbench
0 of 4 ranked benchmarks measured
- AA-Omniscience · Accuracy10.37Oct 8, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination10.21Oct 8, 2026omniscienceNonHallucination
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- MMLU-Pro (HF Open LLM)25.97Oct 8, 2026MMLU-PRO
- 19.94Oct 8, 2026
- 26.59Oct 8, 2026
- MATH Lvl 5 (HF Open LLM)3.85Oct 8, 2026MATH Lvl 5
- 18.39Oct 8, 2026
- IFEval (HF Open LLM)24.01Oct 8, 2026IFEval
Show 54 more resultsHide 54 results
- 0.00Jun 18, 2026
- AA Intelligence4.83Oct 8, 2026aa_intelligence_index
- -70.12Oct 8, 2026
- 42.00Aug 29, 2026
- 70.92May 19, 2026
- 57.30Aug 29, 2026
- 98.83May 10, 2026
- 82.80Aug 29, 2026
- ARC-Challenge59.30Aug 26, 2026ARC-C
- 3.98Jun 18, 2026
- 57.70Aug 26, 2026
- 76.50May 10, 2026
- 51.50Aug 29, 2026
- 80.90Aug 29, 2026
- 49.50Aug 26, 2026
- 50.80Aug 26, 2026
- 40.30Aug 26, 2026
- 49.90Jul 2, 2026
- 56.00Aug 26, 2026
- 77.40Aug 29, 2026
- 72.69May 10, 2026
- 71.10Aug 29, 2026
- 82.10Aug 26, 2026
- 60.40Aug 29, 2026
- 33.50Aug 26, 2026
- livebench_language18.93Aug 23, 2026livebench_language@2025-04-07
- 39.10May 10, 2026
- 41.10Aug 29, 2026
- 20.50Aug 26, 2026
- 67.70Aug 29, 2026
- 53.90Aug 26, 2026
- 60.50Aug 29, 2026
- 60.15May 10, 2026
- 66.50Aug 29, 2026
- 66.60Aug 26, 2026
- 35.40Aug 26, 2026
- 36.30Aug 26, 2026
- 22.60Aug 26, 2026
- 75.39May 10, 2026
- 37.80May 10, 2026
- 76.60Jul 2, 2026
- 82.60Aug 29, 2026
- 75.70Aug 29, 2026
- 59.90Aug 29, 2026
- 99.25May 10, 2026
- 73.90Aug 29, 2026
- 22.10Aug 26, 2026
- 67.70Aug 29, 2026
- TruthfulQA63.20Aug 29, 2026TruthfulQA (MC2)
- 44.00Aug 26, 2026
- 65.00Aug 29, 2026
- 77.40Aug 26, 2026
- 95.61May 10, 2026
- τ²-Bench Telecom (AA run)0.00Oct 8, 2026aa_tau2
Llama 3 Instruct 8B: common questions
Who makes Llama 3 Instruct 8B?
Llama 3 Instruct 8B is made by Meta.
When was Llama 3 Instruct 8B released?
Llama 3 Instruct 8B was released on Apr 18, 2024, according to Artificial Analysis.
How much does Llama 3 Instruct 8B cost?
Llama 3 Instruct 8B costs $0.04 per million input tokens and $0.14 per million output tokens, according to Artificial Analysis. At a mix of three input tokens to one output token, it is cheaper than 94% of the 330 priced models we track.
How many benchmarks has Llama 3 Instruct 8B been tested on?
We track 73 results for Llama 3 Instruct 8B on 59 benchmarks from 17 sources, 21 of them independently verified. The latest was recorded on Oct 8, 2026.
Which API features does Llama 3 Instruct 8B support?
OpenRouter lists structured outputs for Llama 3 Instruct 8B.
About this record
Where Llama 3 Instruct 8B's numbers come from, and every name it appears under.
- Tracked since
- May 2, 2026
- Newest source mention
- Sep 10, 2026
Where the results come from
Verification: 73 scores · 21 independently verified · 14 aggregator-attributed · 38 vendor cross-reference. How these tiers are assigned
From 17 sources on 3 sites. Hugging Face supplies 47 of them; the 21 independently verified results come from 2 sites. Bars are coloured by trust tier.
- huggingface.co47
- artificialanalysis.ai14
- storage.googleapis.com12