Up to date
Qwen Apache 2.0 out 13 months ago
236B parameters, 22B active (8 of 128 experts per token), images and text in, text out, 256K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #155 in LLMs · score 92 method v1 Sources 4 Downloads 30d 60.3k · #366 of 619 tracked
Picked by rule from 18 tracked pieces: the lab's own post, the most shared, a field report, the latest.
OpenAI released gpt-oss-120b and gpt-oss-20b, their first open-weight models since GPT-2 in 2019.
The 3 picks above, plus these 15. 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.
Top Spaces running it: hadadxyz/ai ♥ 819 · umint/ai ♥ 258 · nvidia/Cosmos3-Action-Viewer ♥ 46 · zakaria-ch49/Medical-Question-Answering-with-Retrieval-Augmented-Generation_RAG ♥ 2
Weights only, an estimate: leave room for the KV cache. Method
About 141 GB at INT4, 283 GB at FP8, 566 GB at BF16: weights only, so leave room for the KV cache. Every expert must be resident, so total parameters count.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2025-09-22), amber triangles are 6 dated articles (each links to it), grey ticks are repo commits.
19 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.
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.
| Pipeline tag | image-text-to-text |
|---|---|
| Library | transformers |
| Architectures | Qwen3VLMoeForConditionalGeneration |
| Model type | qwen3_vl_moe |
| Dtype | bfloat16 |
| Params by dtype | BF16: 235,670,022,896 |
| Size band | XXL |
| MoE | yes |
| Experts | 128 |
| Experts per token | 8 |
| Context length | 262,144 (config.json max_position_embeddings) |
| Papers | arXiv:2505.09388, arXiv:2502.13923, arXiv:2409.12191, arXiv:2308.12966 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · Qwen3VLMoeForConditionalGeneration · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen3VLMoeForConditionalGeneration · 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 Qwen3-VL-235B-A22B-Thinking: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 566 GB · Multi-GPU node (<640 GB) |
|---|---|
| FP8 (1 byte/param) | 283 GB · Multi-GPU node (<640 GB) |
| INT4 (0.5 bytes/param) | 141 GB · Multi-GPU node (<640 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra. MoE: all experts must be resident, so total parameters count.
quantized: 15finetune: 3merge: 1
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-VL-235B-A22B-Thinking-FP8 | quantized | -FP8 | 236B | XXL | apache-2.0 | Qwen | 35 | 195.3k |
| unsloth/Qwen3-VL-235B-A22B-Thinking-GGUF | quantized | GGUF | — | — | apache-2.0 | unsloth | 43 | 89k |
| Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF | quantized | GGUF | — | — | apache-2.0 | Qwen | 8 | 35.9k |
| QuantTrio/Qwen3-VL-235B-A22B-Thinking-AWQ | quantized | AWQ | 236B | XXL | apache-2.0 | QuantTrio | 8 | 29k |
| unsloth/Qwen3-VL-235B-A22B-Thinking-1M-GGUF | quantized | GGUF | — | — | apache-2.0 | unsloth | 2 | 27.5k |
| lmstudio-community/Qwen3-VL-235B-A22B-Thinking-GGUF | quantized | GGUF | — | — | apache-2.0 | lmstudio-community | 0 | 11.5k |
| DevQuasar/Qwen.Qwen3-VL-235B-A22B-Thinking-GGUF | quantized | GGUF | — | — | — | DevQuasar | 0 | 5.2k |
| QuantTrio/Qwen3-VL-235B-A22B-Thinking-FP8 | quantized | -FP8 | 236B | XXL | apache-2.0 | QuantTrio | 0 | 4.1k |
| GadflyII/Qwen3-VL-235B-A22B-Thinking-NVFP4 | quantized | NVFP4 | 133B | XXL | apache-2.0 | GadflyII | 0 | 3.2k |
| noctrex/Qwen3-VL-235B-A22B-Thinking-1M-MXFP4_MOE-GGUF | quantized | GGUF | — | — | — | noctrex | 3 | 2.9k |
| Hcompany/Holo2-235B-A22B | finetune | — | 236B | XXL | cc-by-nc-4.0 | Hcompany | 34 | 865 |
| GaleneAI/Qwen3-VL-235B-A22B-Thinking-NVFP4 | quantized | NVFP4 | 133B | XXL | apache-2.0 | GaleneAI | 1 | 1k |
| zenlm/zen-designer-235b-a22b-thinking | finetune | — | 236B | XXL | apache-2.0 | zenlm | 2 | 895 |
| mradermacher/Qwen3-VL-235B-A22B-Thinking-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 2 | 762 |
| unsloth/Qwen3-VL-235B-A22B-Thinking | finetune | — | 236B | XXL | apache-2.0 | unsloth | 8 | 457 |
| mradermacher/Qwen3-VL-235B-A22B-Thinking-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 572 |
| shivak/Qwen3-VL-235B-A22B-Thinking-W4A16 | quantized | -W4A16 | 32.9B | L | apache-2.0 | shivak | 0 | 391 |
| turboderp/Qwen3-VL-235B-A22B-Thinking-exl3 | quantized | exl3 | — | — | apache-2.0 | turboderp | 1 | 217 |
| Abigail45/Smol-DeepSWE | merge | — | — | — | apache-2.0 | Abigail45 | 1 | 0 |
| Date | Commit | SHA |
|---|---|---|
| 2025-11-26 | Adding `transformers` as the library name (#14) | 6664aff |
| 2025-10-04 | Update README.md | d3b3ca4 |
| 2025-09-23 | Update README.md | c6a4947 |
| 2025-09-23 | Update README.md | 6e2d066 |
| 2025-09-23 | Update README.md | df855a8 |
| 2025-09-23 | Update generation_config.json | 6c55e1d |
| 2025-09-23 | Update README.md | 644e941 |
| 2025-09-23 | Update README.md | de1485a |
| 2025-09-23 | Add files using upload-large-folder tool | 460ea5f |
| 2025-09-23 | Add files using upload-large-folder tool | 62ac206 |
| 2025-09-23 | Add files using upload-large-folder tool | 4a698c0 |
| 2025-09-23 | Add files using upload-large-folder tool | 55dc971 |
| 2025-09-23 | Add files using upload-large-folder tool | 65bf4de |
| 2025-09-23 | Add files using upload-large-folder tool | 992342a |
| 2025-09-22 | initial commit | 05f5d44 |
Hosted endpoints observed for Qwen3-VL-235B-A22B-Thinking, 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 | Qwen/Qwen3-VL-235B-A22B-Thinking | — | — | — | — | — | — | — | 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 21:35 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:01 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.
