Up to date
NousResearch Llama 3.1 Community License out 2 years ago
8.03B parameters, text in, text out, 128K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #571 in LLMs · score 74.5 method v1 Sources 3 Downloads 30d 143.2k · #278 of 619 tracked
Picked by rule from 25 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 2 picks above, plus these 23. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | Featherless | — | — | — | — | Oct 25 |
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; 15 quantized builds for local use; a big base for builders (221 fine-tunes, adapters and merges).
Top Spaces running it: Raiff1982/Codette-Reasoning-Demo ♥ 1
Weights only, an estimate: leave room for the KV cache. Method
About 4.8 GB at INT4, 9.6 GB at FP8, 19 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2024-07-24), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.
236 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported AGIEval English | 47.8 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported ARC Challenge · Chat and follow instructions | 79.7 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported BBH · Reason through hard problems | 64.2 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 75 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 77.6 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 75 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported DROP (F1) | 59.5 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported MMLU · Reason through hard problems | 66.7 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported MMLU · Reason through hard problems | 69.4 percent | Instruction tuned models, 8B | model author | 2026-10-07 |
| author-reported MMLU-Pro · Reason through hard problems | 37.1 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported QuAC (F1) | 44.9 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported SQuAD | 77 percent | Base pretrained models, 8B | model author | 2026-10-07 |
| author-reported Winogrande · Chat and follow instructions | 60.5 percent | Base pretrained models, 8B | model author | 2026-10-07 |
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | LlamaForCausalLM |
| Model type | llama |
| Dtype | bfloat16 |
| Params by dtype | BF16: 8,030,261,248 |
| Size band | M |
| MoE | no |
| Context length | 131,072 (config.json max_position_embeddings) |
| Languages | en, de, fr, it, pt, hi, es, th |
| Papers | arXiv:2204.05149 |
| License | Llama 3.1 Community License (llama3.1) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · LlamaForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · LlamaForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for Meta-Llama-3.1-8B-Instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 19 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 9.6 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.8 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 15adapter: 85merge: 5finetune: 131
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-rlhf-dpo-8B | quantized | GGUF | 8.03B | M | apache-2.0 | Saxo | 3 | 26.8k |
| gotzmann/1OV77 | adapter | — | — | — | other | gotzmann | 0 | 9k |
| gotzmann/1JV43 | adapter | — | — | — | other | gotzmann | 0 | 8.7k |
| gotzmann/1JV42II | adapter | — | — | — | — | gotzmann | 0 | 7.4k |
| gotzmann/1JV36 | adapter | — | — | — | other | gotzmann | 0 | 7.1k |
| gotzmann/1JV36B | adapter | — | — | — | other | gotzmann | 0 | 7k |
| gotzmann/1JV32 | adapter | — | — | — | other | gotzmann | 0 | 6.6k |
| Cookieeeeeeeeeeeeeee/llama_custom_zero | adapter | — | — | — | — | Cookieeeeeeeeeeeeeee | 0 | 5.5k |
| llm-wizard/llama381binstruct_summarize | adapter | — | — | — | llama3.1 | llm-wizard | 0 | 5.5k |
| gotzmann/1SV64 | adapter | — | — | — | other | gotzmann | 0 | 5.2k |
| gotzmann/1JV44C2nd | adapter | — | — | — | other | gotzmann | 0 | 5.2k |
| gotzmann/1JV45FSDP | adapter | — | — | — | other | gotzmann | 0 | 4.8k |
| yinong333/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | yinong333 | 0 | 4.7k |
| gotzmann/1JV42 | adapter | — | — | — | other | gotzmann | 0 | 4.2k |
| gotzmann/1JV35 | adapter | — | — | — | other | gotzmann | 0 | 4.1k |
| rchrdgwr/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | rchrdgwr | 0 | 4k |
| pattonma/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | pattonma | 0 | 3.9k |
| llm-wizard/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | llm-wizard | 0 | 3.9k |
| BaselMousi/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | BaselMousi | 0 | 3.9k |
| jimmydzj2006/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | jimmydzj2006 | 0 | 3.9k |
| Galatea007/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | Galatea007 | 0 | 3.9k |
| ekinnk/checkpoint-AT-atamalar-10k | adapter | — | — | — | — | ekinnk | 0 | 3.8k |
| Cheselle/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | Cheselle | 0 | 3.8k |
| achapman/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | achapman | 0 | 3.6k |
| gotzmann/1SV47 | adapter | — | — | — | other | gotzmann | 0 | 3.6k |
| acpotts/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | acpotts | 0 | 3.5k |
| gotzmann/1JV40 | adapter | — | — | — | other | gotzmann | 0 | 3.4k |
| jet-taekyo/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | jet-taekyo | 0 | 3.4k |
| niting089/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | niting089 | 0 | 3.2k |
| gotzmann/1SV71 | adapter | — | — | — | other | gotzmann | 0 | 3.2k |
| skybarley/llama_custom_zero | adapter | — | — | — | — | skybarley | 0 | 3.2k |
| dgutierrez/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | dgutierrez | 0 | 3.1k |
| gotzmann/1SV52 | adapter | — | — | — | other | gotzmann | 0 | 2.8k |
| featherless-ai-quants/NousResearch-Meta-Llama-3.1-8B-Instruct-GGUF | quantized | GGUF | — | — | — | featherless-ai-quants | 0 | 2.1k |
| gotzmann/1OV84 | adapter | — | — | — | other | gotzmann | 0 | 1.9k |
| liodon-ai/Meta-Llama-3.1-8B-Instruct-imatrix-GGUF | quantized | imatrix | — | — | other | liodon-ai | 0 | 1.6k |
| XicoC/llama381binstruct_summarize_short | adapter | — | — | — | llama3.1 | XicoC | 0 | 1.5k |
| gotzmann/1JV44C | adapter | — | — | — | other | gotzmann | 0 | 832 |
| gotzmann/1SV48FULL | adapter | — | — | — | other | gotzmann | 0 | 819 |
| gotzmann/1JV45 | adapter | — | — | — | other | gotzmann | 0 | 692 |
Hosted endpoints observed for Meta-Llama-3.1-8B-Instruct, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| Featherless hf | NousResearch/Meta-Llama-3.1-8B-Instruct | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency.
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:40 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:12 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
