The story so far
Weights, every dated article, and repo commits on one rail.
18 repo commits since the weights landed
Teal is the weights (2025-04-10), grey ticks are repo commits.
InternVL3-1B
OpenGVLab Apache 2.0 (qwen) out 18 months ago
938M parameters, images and text in, text out.
Verified 2026-10-25 Rank #706 in LLMs · score 72 method v1 Sources 3 Downloads 30d 69k · #358 of 619 tracked
Do people use it?69kdownloads, 30 days#358 of 619 we track · below the median85 likesHugging Face2026-10-24
Where it's used
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.
No API provider serves it, so its downloads are people running it themselves: 6 quantized builds and 2 Spaces.
- As a service0no API provider listed
- On own hardware6quantized builds
- Built on139 fine-tunes · 4 adapters · 0 merges
- In apps2Hugging Face Spaces
Top Spaces running it: TSXu/UniCalli_Dev ♥ 7 · zaleni/wsa1-robot-control ♥ 1
Will it fit?
Weights only, an estimate: leave room for the KV cache. Method
Fits a 16 GB card at INT4
About 0.6 GB at INT4, 1.1 GB at FP8, 2.3 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
What people built on it
19 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
19 derivatives on the Hub, led by 9 fine-tunes
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Reference
Everything else, folded: the full tables, every source.
Use-case evidence
6 use cases
What it is good at 6 use cases
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.
- Understand images and videostructuraldecision-model estimate p=1.00tag: image-text-to-textpipeline: image-text-to-text+1 more
- Chat and follow instructionsstructuraldecision-model estimate p=0.98tag: conversationalchat template present
- Run on a laptop or phonestructuraldecision-model estimate p=1.00name: 1Bfits: edge
- Embed text for searchstructuraltag: feature-extraction
- Work across languagesstructuraltag: multilingual
- Call tools and APIsstructuraldecision-model estimate p=0.81
Facts
19 fields
Listed in LLMs
Facts
- InternVL3-1B has 938M parameters. (safetensors metadata, observed 2026-10-24)
- License: Apache 2.0 (qwen) (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
- Created 2025-04-10, last modified 2025-09-11. (HF API, observed 2026-10-24)
| Pipeline tag | image-text-to-text |
|---|---|
| Library | transformers |
| Architectures | InternVLChatModel |
| Model type | internvl_chat |
| Dtype | bfloat16 |
| Params by dtype | BF16: 938,193,024 |
| Size band | S |
| MoE | no |
| Languages | multilingual |
| Papers | arXiv:2312.14238, arXiv:2404.16821, arXiv:2412.05271, arXiv:2411.10442, arXiv:2504.10479, arXiv:2412.09616 |
| License | Apache 2.0 (qwen) (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Servability
vLLM · SGLang
Servability
vLLM: supported · InternVLChatModel · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · InternVLChatModel · 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.
Hardware fit
3 estimates
Hardware fit ESTIMATE
Weights-only memory for InternVL3-1B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 2.3 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 1.1 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.6 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
All variants
19 variants
Lineage
Base model: OpenGVLab/InternVL3-1B-Instruct
finetune: 9quantized: 6adapter: 4
Top variants (19 shown, ranked by downloads)
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| unsloth/InternVL3-1B-GGUF | finetune | GGUF | — | — | apache-2.0 | unsloth | 7 | 12.6k |
| mradermacher/InternVL3-1B-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 9.4k |
| mradermacher/InternVL3-1B-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 1 | 6.1k |
| OpenGVLab/InternVL3-1B-AWQ | quantized | AWQ | — | — | other | OpenGVLab | 2 | 2.7k |
| unsloth/InternVL3-1B | finetune | — | 938M | S | apache-2.0 | unsloth | 0 | 2.7k |
| kanashi6/UniLIP-1B | finetune | — | 1.79B | S | apache-2.0 | kanashi6 | 3 | 924 |
| litert-community/InternVL3-1B | quantized | — | — | — | apache-2.0 | litert-community | 0 | 759 |
| AdaptLLM/biomed-InternVL3-1B | finetune | — | 938M | S | other | AdaptLLM | 1 | 507 |
| AXERA-TECH/InternVL3-1B | quantized | — | — | — | bsd-3-clause | AXERA-TECH | 1 | 407 |
| MANO066/Intervl3-1b-safety | finetune | — | 938M | S | apache-2.0 | MANO066 | 0 | 295 |
| D-Robotics/InternVL3-1B-GGUF-BPU | quantized | GGUF | — | — | apache-2.0 | D-Robotics | 0 | 141 |
| kanashi6/UniLIP | finetune | — | — | — | apache-2.0 | kanashi6 | 2 | 0 |
| Yukino271828/SoliReward | finetune | — | — | — | mit | Yukino271828 | 2 | 0 |
| Kablo/UniLIP-1B | finetune | — | 1.79B | S | apache-2.0 | Kablo | 1 | 45 |
| blind-assist/internvl3-1b-walk-lora-Epoch3-8500-v1 | adapter | — | — | — | apache-2.0 | blind-assist | 0 | 41 |
| blind-assist/internvl3-1b-walk-lora-Epoch3-8500-without_ES_v1 | adapter | — | — | — | apache-2.0 | blind-assist | 0 | 40 |
| blind-assist/internvl3-1b-walk-lora-Epoch_1.85-8500-without_ES_v1 | adapter | — | — | — | apache-2.0 | blind-assist | 0 | 37 |
| quangnd58/evo1-so101-multitask | finetune | — | 775M | S | apache-2.0 | quangnd58 | 0 | 30 |
| blind-assist/internvl3-1b-walk-v1-full-checkpoint | adapter | — | 1.01B | S | — | blind-assist | 0 | 19 |
Repository changelog
18 commits
Repository changelog 18 commits
| Date | Commit | SHA |
|---|---|---|
| 2025-09-11 | Upload tokenizer_config.json with huggingface_hub | 4415a3b |
| 2025-09-11 | Upload tokenizer_config.json with huggingface_hub | aed473f |
| 2025-05-29 | Update README.md | 06cddba |
| 2025-04-25 | Update README.md | e5092de |
| 2025-04-17 | Upload README.md with huggingface_hub | b7f05db |
| 2025-04-16 | Upload modeling_intern_vit.py with huggingface_hub | 4533f46 |
| 2025-04-15 | Upload modeling_internvl_chat.py with huggingface_hub | 6467c5d |
| 2025-04-15 | Upload README.md with huggingface_hub | 2c408ff |
| 2025-04-14 | Upload README.md with huggingface_hub | 0fb4585 |
| 2025-04-12 | Upload README.md with huggingface_hub | 40e62a4 |
| 2025-04-12 | Upload README.md with huggingface_hub | 9a6ecab |
| 2025-04-12 | Upload README.md with huggingface_hub | 15c857b |
| 2025-04-12 | Upload README.md with huggingface_hub | bf60797 |
| 2025-04-10 | Upload folder using huggingface_hub | 4014eae |
| 2025-04-10 | Upload folder using huggingface_hub | 1c77635 |
| 2025-04-10 | Upload folder using huggingface_hub | a8221dd |
| 2025-04-10 | Upload folder using huggingface_hub | 698c83f |
| 2025-04-10 | initial commit | 1dabde7 |
Papers
6 papers
Papers 6
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).
- V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding cited in card peer-reviewed · measuredJunqi Ge, Ziyi Chen, Jintao Lin +4 · 2024-12-12 · IEEE/CVF International Conference on Computer Vision (ICCV) · PDF · HF paper page (1 upvotes) · evidence
Similar models
3 models
Similar models
- Qwen3.5-0.8B · 873M same size band (S), same task (image-text-to-text), more providers, permissive license
- LightOnOCR-2-1B · 1.01B same size band (S), same task (image-text-to-text), more providers, permissive license
- NVIDIA-Nemotron-Parse-v1.1 · 957M same size band (S), same task (image-text-to-text)
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:43 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:16 UTC · Last verified 2026-10-25
Follow InternVL3-1B
Watch this model
- /feeds/models/OpenGVLab--InternVL3-1B.xml(RSS 2.0) · every change to InternVL3-1B
- /feeds/families/OpenGVLab--InternVL3.xml(RSS 2.0) · every change across the InternVL3 family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.