Up to date
lmsys Llama 2 Community License out 3 years ago
Text in, text out, 4K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #349 in LLMs · score 81.3 method v1 Sources 3 Downloads 30d 97.3k · #318 of 619 tracked
Picked by rule from 6 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The lmsys/vicuna-7b-v1.5 model ranks 32 out of 412 on the ShoppingMMLU benchmark with an average score of 45.9%.
The 2 picks above, plus these 4. 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; 24 quantized builds for local use; a big base for builders (268 fine-tunes, adapters and merges).
Top Spaces running it: h2oai/h2ogpt-chatbot ♥ 295 · h2oai/h2ogpt-chatbot2 ♥ 167 · lixin4ever/VideoLLaMA2 ♥ 160 · shi-labs/VCoder ♥ 63
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2023-07-29), amber triangles are 1 dated article (each links to it), grey ticks are repo commits.
292 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.
Vicuna-7b-v1.5 is an auto-regressive language model based on the transformer architecture.[1] Developed by LMSYS, this model is fine-tuned from Llama 2 for research on large language models and chatbots.[2][3]
Generated 2026-10-07 15:06 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
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 | float16 |
| MoE | no |
| Context length | 4,096 (config.json max_position_embeddings) |
| Papers | arXiv:2307.09288, arXiv:2306.05685 |
| License | Llama 2 Community License (llama2) |
| 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.
merge: 12quantized: 24finetune: 70adapter: 186
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| X-iZhang/libra-llava-rad | merge | — | 6.76B | M | — | X-iZhang | 2 | 91.8k |
| TheBloke/vicuna-7B-v1.5-GPTQ | quantized | GPTQ | 6.74B | M | llama2 | TheBloke | 17 | 78.4k |
| X-iZhang/libra-maira-2 | merge | — | 6.88B | M | — | X-iZhang | 3 | 71.3k |
| TheBloke/vicuna-7B-v1.5-GGUF | quantized | GGUF | — | — | llama2 | TheBloke | 16 | 44.5k |
| OpenGVLab/InternVL-Chat-ViT-6B-Vicuna-7B | merge | — | — | — | — | OpenGVLab | 8 | 4.5k |
| TheBloke/vicuna-7B-v1.5-AWQ | quantized | AWQ | 6.74B | M | llama2 | TheBloke | 4 | 4.5k |
| chaubeyG/FaceLLaVA | finetune | — | — | — | other | chaubeyG | 3 | 3.9k |
| mradermacher/vicuna-7b-v1.5-i1-GGUF | quantized | GGUF | — | — | llama2 | mradermacher | 0 | 3.5k |
| mradermacher/vicuna-7b-v1.5-GGUF | quantized | GGUF | — | — | llama2 | mradermacher | 0 | 2.2k |
| TheBloke/vicuna-7B-v1.5-GGML | finetune | — | — | — | llama2 | TheBloke | 14 | 550 |
| YipengZhang/LLaVA-UHD-v2-Vicuna-7B | finetune | — | 7.52B | M | — | YipengZhang | 6 | 798 |
| wentao-yuan/robopoint-v1-vicuna-v1.5-7b-lora | finetune | — | — | — | apache-2.0 | wentao-yuan | 0 | 1k |
| tsunghanwu/reverse_llava_v15 | finetune | — | 6.76B | M | mit | tsunghanwu | 1 | 824 |
| JoshuaW1997/FUTGA | finetune | — | — | — | apache-2.0 | JoshuaW1997 | 17 | 0 |
| LeyaGao/vicuna-7b-v1.5-Q4_K_M-GGUF | quantized | GGUF | — | — | llama2 | LeyaGao | 0 | 617 |
| Charlie911/vicuna-7b-v1.5-lora-mctaco | adapter | — | — | — | — | Charlie911 | 0 | 578 |
| shikiw/LLaVA-v1.5-MoCa-7B | finetune | — | 7.06B | M | llama2 | shikiw | 2 | 469 |
| TheMarmot/vicuna-7b-v1.5-gguf | quantized | gguf | — | — | llama2 | TheMarmot | 3 | 400 |
| MCG-NJU/p-MoD-LLaVA-v1.5-7B | finetune | — | 7.06B | M | apache-2.0 | MCG-NJU | 3 | 308 |
| laislemke/LLaMA-2-vicuna-7b-slerp | merge | — | 6.74B | M | llama2 | laislemke | 0 | 440 |
| zzachw12/llemr-v1 | adapter | — | — | — | mit | zzachw12 | 1 | 383 |
| shikiw/LLaVA-v1.5-MoCa-7B-pretrain | finetune | — | 7.06B | M | llama2 | shikiw | 1 | 304 |
| OpenGVLab/PIIP-LLaVA_ConvNeXt-B_CLIP-L_640-224_7B | merge | — | 7.18B | M | mit | OpenGVLab | 0 | 338 |
| OpenGVLab/PIIP-LLaVA_CLIP-BL_512-448_7B | merge | — | 7.22B | M | mit | OpenGVLab | 0 | 322 |
| OpenGVLab/PIIP-LLaVA-Plus_ConvNeXt-L_CLIP-L_1024-336_7B | merge | — | 7.34B | M | mit | OpenGVLab | 0 | 319 |
| OpenGVLab/PIIP-LLaVA_ConvNeXt-L_CLIP-L_1024-336_7B | merge | — | 7.34B | M | mit | OpenGVLab | 0 | 318 |
| BASH-Lab/LLaSA-7B | finetune | — | 6.76B | M | other | BASH-Lab | 0 | 316 |
| MCG-NJU/p-MoD-LLaVA-NeXT-7B | finetune | — | 7.06B | M | apache-2.0 | MCG-NJU | 4 | 114 |
| OpenGVLab/PIIP-LLaVA_CLIP-BL_512-256_7B | merge | — | 7.18B | M | mit | OpenGVLab | 0 | 310 |
| OpenGVLab/PIIP-LLaVA_ConvNeXt-B_CLIP-L_1024-336_7B | merge | — | 7.21B | M | mit | OpenGVLab | 0 | 299 |
| joehuangx/spatial-vicuna-7b-v1.5-LoRA | adapter | — | — | — | — | joehuangx | 1 | 200 |
| Fabio2000/vicuna-7b-v1.5-Q4_K_M-GGUF | quantized | GGUF | — | — | llama2 | Fabio2000 | 0 | 244 |
| gwkrsrch/MambaMia-Vicuna-7B-Mini | finetune | — | 7.40B | M | llama2 | gwkrsrch | 0 | 231 |
| puar-playground/TextLap-COCO | finetune | — | — | — | mit | puar-playground | 0 | 230 |
| JosephTong/llava-v1.5-7b-flowcut128 | finetune | — | 7.06B | M | apache-2.0 | JosephTong | 1 | 167 |
| itlwas/vicuna-7b-v1.5-Q4_K_M-GGUF | quantized | GGUF | — | — | llama2 | itlwas | 0 | 213 |
| puar-playground/TextLap-Graphic | finetune | — | — | — | mit | puar-playground | 1 | 154 |
| tt1225/aic24-track2-multiview-videollava-7b-lora | adapter | — | — | — | — | tt1225 | 0 | 189 |
| lavinal712/transfusion-7b | finetune | — | 13.2B | L | mit | lavinal712 | 1 | 134 |
| preetam7/ClaimVer_Vicuna1.5-7B-Chat | adapter | — | — | — | llama2 | preetam7 | 1 | 117 |
| Date | Commit | SHA |
|---|---|---|
| 2024-03-13 | Revert chat template | 3321f76 |
| 2024-03-12 | Add chat_template in the tokenizer (#7) | 90cea65 |
| 2023-08-02 | Update README.md | de56c35 |
| 2023-08-01 | Update README.md | 5070998 |
| 2023-08-01 | Update README.md | dbb9c3d |
| 2023-08-01 | Update README.md | ab33688 |
| 2023-08-01 | Update README.md | 8f10e48 |
| 2023-07-30 | Update README.md | aea9af8 |
| 2023-07-30 | Update README.md | f5b48ee |
| 2023-07-30 | Update README.md | c855bfe |
| 2023-07-30 | Create README.md | 14209c0 |
| 2023-07-29 | Upload tokenizer | 73fb5df |
| 2023-07-29 | Upload LlamaForCausalLM | c1b584e |
| 2023-07-29 | initial commit | 25ba69c |
Hosted endpoints observed for vicuna-7b-v1.5, 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 | lmsys/vicuna-7b-v1.5 | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
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 21:37 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:06 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.
