Up to date
Qwen Apache 2.0 out 2 years ago
32.8B parameters, text in, text out, 128K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #319 in LLMs · score 82.4 method v1 Sources 3 Downloads 30d 19.7k · #458 of 619 tracked
Picked by rule from 26 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 1 pick above, plus these 24. 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; heavily run locally (72 quantized builds); a big base for builders (251 fine-tunes, adapters and merges).
Top Spaces running it: ISEEKYAN/megatron_memory_estimator ♥ 35 · ISEEKYAN/megatron_memory_estimator_old ♥ 19 · artificialguybr/CanIRunThisLLM ♥ 9 · CATIE-AQ/Guide_Evaluation_LLM ♥ 5
Weights only, an estimate: leave room for the KV cache. Method
About 20 GB at INT4, 39 GB at FP8, 79 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 (2024-09-15), amber triangles are 3 dated articles (each links to it), grey ticks are repo commits.
323 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.
It supports a context length of 131,072 tokens.[1]
Generated 2026-10-07 15:09 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 |
|---|---|
| Architectures | Qwen2ForCausalLM |
| Model type | qwen2 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 32,763,876,352 |
| Size band | L |
| MoE | no |
| Context length | 131,072 (config.json max_position_embeddings) |
| Languages | en |
| Papers | arXiv:2407.10671 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · Qwen2ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen2ForCausalLM · 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 Qwen2.5-32B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 79 GB · Single high-end GPU (<80 GB) |
|---|---|
| FP8 (1 byte/param) | 39 GB · Single high-end GPU (<80 GB) |
| INT4 (0.5 bytes/param) | 20 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
finetune: 128quantized: 72merge: 66adapter: 57
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| baichuan-inc/Baichuan-M2-32B | finetune | — | 32.8B | L | apache-2.0 | baichuan-inc | 127 | 1.2M |
| unsloth/Qwen2.5-32B-bnb-4bit | quantized | -bnb | 33.7B | L | apache-2.0 | unsloth | 3 | 483.2k |
| deepcogito/cogito-v1-preview-qwen-32B | finetune | — | 32.8B | L | apache-2.0 | deepcogito | 117 | 319.7k |
| mlx-community/Qwen2.5-32B-Instruct-4bit | finetune | -4bit | 32.8B | L | apache-2.0 | mlx-community | 4 | 302.3k |
| nvidia/OpenReasoning-Nemotron-32B | finetune | — | 32.8B | L | cc-by-4.0 | nvidia | 128 | 252k |
| CalamitousFelicitousness/Qwen2.5-32B-Instruct-fp8-dynamic | quantized | -fp8 | 32.8B | L | apache-2.0 | CalamitousFelicitousness | 2 | 213.9k |
| IFM/K2-Think | finetune | — | 32.8B | L | apache-2.0 | IFM | 369 | 122.3k |
| BytedTsinghua-SIA/DAPO-Qwen-32B | finetune | — | 32.8B | L | apache-2.0 | BytedTsinghua-SIA | 12 | 37.3k |
| unsloth/Qwen2.5-32B | finetune | — | 32.8B | L | apache-2.0 | unsloth | 2 | 30k |
| RefalMachine/RuadaptQwen2.5-32B-Pro-Beta | finetune | — | 32.7B | L | apache-2.0 | RefalMachine | 13 | 27.4k |
| mradermacher/Qwen2.5-32B-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 27.7k |
| mradermacher/Qwen2.5-32B-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 1 | 27.5k |
| Mungert/QwQ-32B-GGUF | quantized | GGUF | — | — | apache-2.0 | Mungert | 20 | 21.6k |
| arcee-ai/Virtuoso-Medium-v2-GGUF | quantized | GGUF | — | — | apache-2.0 | arcee-ai | 12 | 21.2k |
| ThomasBaruzier/Qwen2.5-32B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | ThomasBaruzier | 1 | 21.1k |
| EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2 | finetune | — | 32.8B | L | apache-2.0 | EVA-UNIT-01 | 65 | 17.8k |
| trashpanda-org/QwQ-32B-Snowdrop-v0 | merge | — | 32.8B | L | — | trashpanda-org | 104 | 13k |
| rinna/deepseek-r1-distill-qwen2.5-bakeneko-32b | merge | — | 32.8B | L | apache-2.0 | rinna | 7 | 15.9k |
| Mungert/LongWriter-Zero-32B-GGUF | quantized | GGUF | — | — | apache-2.0 | Mungert | 2 | 13.3k |
| arcee-ai/Virtuoso-Medium-v2 | finetune | — | 32.8B | L | apache-2.0 | arcee-ai | 58 | 9.6k |
| THU-KEG/LongWriter-Zero-32B | finetune | — | 32.8B | L | apache-2.0 | THU-KEG | 115 | 6k |
| rinna/qwen2.5-bakeneko-32b-instruct | merge | — | 32.8B | L | apache-2.0 | rinna | 7 | 10.9k |
| nvidia/OpenMath-Nemotron-32B | finetune | — | 32.8B | L | cc-by-4.0 | nvidia | 35 | 9.3k |
| MiniMaxAI/SynLogic-Mix-3-32B | finetune | — | 32.8B | L | mit | MiniMaxAI | 20 | 9.9k |
| Mungert/OpenMath-Nemotron-32B-GGUF | quantized | GGUF | — | — | cc-by-4.0 | Mungert | 1 | 10.1k |
| baichuan-inc/Baichuan-M2-32B-GPTQ-Int4 | quantized | GPTQ | 32.8B | L | apache-2.0 | baichuan-inc | 12 | 9.2k |
| baichuan-inc/Baichuan-M2-32B-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | baichuan-inc | 3 | 9.4k |
| Kwaipilot/SRPO-Qwen-32B | finetune | — | 32.8B | L | mit | Kwaipilot | 16 | 8.5k |
| clowman/QwQ-32B-GPTQ-Int4 | quantized | GPTQ | 32.8B | L | apache-2.0 | clowman | 0 | 9.1k |
| rinna/qwq-bakeneko-32b | merge | — | 32.8B | L | apache-2.0 | rinna | 21 | 7.7k |
| RefalMachine/RuadaptQwen2.5-32B-Pro-Beta-GGUF | quantized | GGUF | — | — | apache-2.0 | RefalMachine | 12 | 5.6k |
| EVA-UNIT-01/EVA-Qwen2.5-32B-v0.0 | finetune | — | 32.8B | L | apache-2.0 | EVA-UNIT-01 | 26 | 4.6k |
| EVA-UNIT-01/EVA-Qwen2.5-32B-v0.1 | finetune | — | 32.8B | L | apache-2.0 | EVA-UNIT-01 | 15 | 4.7k |
| clowman/QwQ-32B-GPTQ-Int8 | quantized | GPTQ | 32.8B | L | apache-2.0 | clowman | 0 | 4.4k |
| mlx-community/Qwen2.5-32B-Instruct-8bit | finetune | -8bit | 32.8B | L | apache-2.0 | mlx-community | 7 | 3.6k |
| MiniMaxAI/SynLogic-32B | finetune | — | 32.8B | L | mit | MiniMaxAI | 17 | 2.8k |
| Naozumi0512/DeepSeek-R1-Distill-Qwen-32B-lora-r32 | finetune | — | — | — | — | Naozumi0512 | 2 | 3.4k |
| flux-inc/Flux-Japanese-Qwen2.5-32B-Instruct-V1.0 | finetune | — | 32.8B | L | apache-2.0 | flux-inc | 10 | 3k |
| internlm/OREAL-32B-SFT | finetune | — | 32.8B | L | apache-2.0 | internlm | 7 | 3k |
| ByteDance-Seed/BFS-Prover-V2-32B | finetune | — | 32.8B | L | apache-2.0 | ByteDance-Seed | 12 | 2.6k |
| Date | Commit | SHA |
|---|---|---|
| 2024-09-20 | Upload ./tokenizer_config.json with huggingface_hub | 1818d35 |
| 2024-09-19 | Upload ./tokenizer_config.json with huggingface_hub | ff23665 |
| 2024-09-18 | update README & LICENSE | 019d2db |
| 2024-09-18 | Update README.md | 56837f7 |
| 2024-09-18 | Create LICENSE | 7ff5cd4 |
| 2024-09-17 | Update README.md | 756235c |
| 2024-09-17 | Update README.md | af53892 |
| 2024-09-17 | Create README.md | 5a17265 |
| 2024-09-16 | Upload tokenizer_config.json with huggingface_hub | bb0be4c |
| 2024-09-15 | Upload folder using huggingface_hub | dbbc0e5 |
| 2024-09-15 | initial commit | 071a8a9 |
Hosted endpoints observed for Qwen2.5-32B, 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/Qwen2.5-32B | — | — | — | — | — | — | — | 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 21:38 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:09 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.
