MetaLlama 3-8B
Llama 3-8B is a Meta open-weights model, released 18 Apr 2024. Its best configuration ranks #764 of 785 on the BenchLeader Index at 35.1 ±3.0, in the lower half. The ± is the point: 114 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (36) and lowest in long context (23). At $0.073 per million tokens blended it is among the cheapest fifth of ranked models. Last measured 22 Oct 2026.
- Blended price
- $0.073/M
- $0.050 in · $0.140 out
- Output speed
- –
- First answer
- –
- Context
- 8k
- Full answer
- –
- Cost per run
- –
- Released
- 18 Apr 2024
- Overall index35
- Reasoning32
- Coding35
- Agents & tools34
- Maths30
- Knowledge30
- Instruction following29
- Human preference36
- Long context23
- Composite35
Versions
Meta has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Llama 3.1 8B | 23 Jul 2024 | 36.1 | #753 |
| Llama 3-8Bthis page | 18 Apr 2024 | 35.1 | #764 |
How its index has moved
8 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| GPQA Diamond | 26.1%#282 | Epoch AI Benchmarking Hub | |
| Arena Hard Prompts | 1220#309 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 29.6%#524 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 5.1%#450 | Artificial Analysis | |
| CritPt | 0.0%#315 | Artificial Analysis | |
| DTBenchnot in index | 43.9%#226 | Epoch AI Benchmarking Hub | |
| ForecastBenchnot in index | 59.4%#73 | ForecastBench | |
| WinoGrandenot in index | 75.7%#32 | Epoch AI Benchmarking Hub | |
| ARC (AI2 Challenge)not in index | 82.8%#16 | Epoch AI Benchmarking Hub | |
| ANLInot in index | 57.3%#3 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1253#304 | Arena (LMArena) |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 0.0%#362 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 0.0%#393 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| OTIS Mock AIME | 1.9%#260 | Epoch AI Benchmarking Hub | |
| MATH Level 5 | 6.1%#92 | Epoch AI Benchmarking Hub | |
| Arena Maths | 1193#304 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -70.1#515 | Artificial Analysis | |
| MMLUnot in index | 68.8%#59 | Epoch AI Benchmarking Hub | |
| TriviaQAnot in index | 67.7%#30 | Epoch AI Benchmarking Hub | |
| OpenBookQAnot in index | 82.6%#6 | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in index | 10.4%#492 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 10.2%#414 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 24.6%#396 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1193#306 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1224#300 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1196#295 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1206#294 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1209#306 | Arena (LMArena) |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 0.0%#493 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 116.4#190 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 4.8#618 | Artificial Analysis |
Price history
The price above, per 1M tokens, every day it changed.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | – | – |
| Summarize a 30-page report | 12,000 / 600 | $0.0007 | – | – |
| Code edit | 6,000 / 1,500 | $0.0005 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.0036 | – | – |
| Structured extraction | 2,000 / 200 | $0.0001 | – | – |
See also
Questions people ask
- What are Llama 3-8B's benchmark scores?
- As of 24 Oct 2026, Llama 3-8B scores 35.1 ±3.0 on the BenchLeader Index, strongest in human preference (36) and weakest in long context (23). That rests on 13 independent benchmark results, and every score on this page links to its publisher.
- How does Llama 3-8B rank?
- Llama 3-8B is #764 of 785 ranked configurations on 24 Oct 2026. 114 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Llama 3-8B cost per 1M tokens?
- As of 24 Oct 2026, Llama 3-8B costs $0.050 per 1M input tokens and $0.140 per 1M output tokens, $0.073 blended (three input tokens per output token). The list price last changed on 9 Sept 2026.
- When was Llama 3-8B released?
- Meta released Llama 3-8B on 18 Apr 2024.
Data as of 24 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Llama 3-8B: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/llama-3-8b.