Up to date
Qwen Apache 2.0 out 2 years ago
7.62B parameters, text in, text out, 4K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #441 in LLMs · score 78.2 method v1 Sources 3 Downloads 30d 47.1k · #387 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; heavily run locally (50 quantized builds); a big base for builders (538 fine-tunes, adapters and merges).
Top Spaces running it: allenai/reward-bench ♥ 438 · BoltzmannEntropy/QuantumLLMInstruct ♥ 7 · Speedofmastery/orynxml-backend ♥ 2 · Shasidharyadavr/netweaver_sre ♥ 2
Weights only, an estimate: leave room for the KV cache. Method
About 4.6 GB at INT4, 9.1 GB at FP8, 18 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 (2024-09-19), amber triangles are 5 dated articles (each links to it), grey ticks are repo commits.
588 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.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | Qwen2ForCausalLM |
| Model type | qwen2 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 7,615,616,512 |
| Size band | M |
| MoE | no |
| Context length | 4,096 (config.json max_position_embeddings) |
| Languages | en |
| Papers | arXiv:2409.12122 |
| 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-Math-7B-Instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 18 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 9.1 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.6 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Base model: Qwen/Qwen2.5-Math-7B
finetune: 141quantized: 50merge: 74adapter: 323
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Qwen/Qwen2.5-Math-PRM-7B | finetune | — | 7.63B | M | other | Qwen | 97 | 2.5M |
| mlx-community/Qwen2.5-Math-7B-Instruct-4bit | quantized | -4bit | 7.62B | M | apache-2.0 | mlx-community | 0 | 246.6k |
| bartowski/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | bartowski | 15 | 238.3k |
| AI-MO/Kimina-Prover-Preview-Distill-7B | finetune | — | 7.62B | M | apache-2.0 | AI-MO | 36 | 55.2k |
| Skywork/Skywork-o1-Open-PRM-Qwen-2.5-7B | finetune | — | — | — | other | Skywork | 53 | 50.4k |
| DrNicefellow/Qwen2.5-Math-7B-Instruct-4.0bpw-exl2 | quantized | 0bpw | — | — | apache-2.0 | DrNicefellow | 1 | 50.4k |
| unsloth/Qwen2.5-Math-7B-Instruct-bnb-4bit | quantized | -bnb | 7.82B | M | apache-2.0 | unsloth | 4 | 49.2k |
| lmstudio-community/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | lmstudio-community | 4 | 32.7k |
| unsloth/Qwen2.5-Math-7B-Instruct | finetune | — | 7.62B | M | apache-2.0 | unsloth | 2 | 29.9k |
| Qwen/Qwen2.5-Math-7B-PRM800K | finetune | — | 7.63B | M | other | Qwen | 27 | 21.7k |
| legraphista/Qwen2.5-Math-7B-Instruct-IMat-GGUF | quantized | GGUF | — | — | apache-2.0 | legraphista | 0 | 18.9k |
| mradermacher/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 16.7k |
| QuantFactory/SuperCorrect-7B-GGUF | quantized | GGUF | — | — | apache-2.0 | QuantFactory | 2 | 11.3k |
| QuantFactory/Qwen2.5-7B-Instruct-MathCoder-GGUF | merge | GGUF | — | — | — | QuantFactory | 4 | 7.9k |
| mradermacher/Qwen2.5-Math-7B-Instruct-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 7.7k |
| RichardErkhov/BitStarWalkin_-_SuperCorrect-7B-gguf | quantized | gguf | — | — | apache-2.0 | RichardErkhov | 0 | 5.6k |
| second-state/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | second-state | 0 | 5.3k |
| mergekit-community/Qwen3-7B-Instruct | merge | — | 7.62B | M | — | mergekit-community | 2 | 4.5k |
| gaianet/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | gaianet | 0 | 3.3k |
| tensorblock/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | apache-2.0 | tensorblock | 0 | 2.9k |
| Yuuta208/Qwen2.5-7B-Instruct-Qwen2.5-Math-7B-Instruct-Merged-ties-29 | merge | — | 7.62B | M | — | Yuuta208 | 0 | 2.7k |
| noirchan/Qwen2.5-Coder-7B_math_mergeTIES | merge | — | 7.62B | M | — | noirchan | 0 | 2.5k |
| DevQuasar-8/Qwen.Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | — | DevQuasar-8 | 0 | 2.3k |
| Locutusque/StockQwen-2.5-7B | merge | — | 7.62B | M | — | Locutusque | 6 | 1.6k |
| MaziyarPanahi/Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | — | MaziyarPanahi | 0 | 1.8k |
| BitStarWalkin/SuperCorrect-7B | finetune | — | 7.62B | M | apache-2.0 | BitStarWalkin | 11 | 1.2k |
| featherless-ai-quants/Qwen-Qwen2.5-Math-7B-Instruct-GGUF | quantized | GGUF | — | — | — | featherless-ai-quants | 0 | 1.7k |
| chiakelvin/Qwen2.5-Math-7B-Instruct-Q4_K_L.gguf | quantized | gguf | — | — | apache-2.0 | chiakelvin | 0 | 1.3k |
| chiakelvin/Qwen2.5-Math-7B-Instruct-Q5_K_M.gguf | quantized | gguf | — | — | apache-2.0 | chiakelvin | 0 | 947 |
| drawais/Qwen2.5-Math-7B-Instruct-AWQ-INT4 | quantized | AWQ | 7.67B | M | apache-2.0 | drawais | 0 | 935 |
| AbteeXAILab/lumynax-math-qwen25-math-7b-gguf | quantized | gguf | — | — | apache-2.0 | AbteeXAILab | 0 | 922 |
| declare-lab/PathFinder-PRM-7B | finetune | — | 7.62B | M | mit | declare-lab | 5 | 606 |
| ZaandaTeika/Qwen2.5-Math-7B-SHARP-Span | finetune | — | 7.08B | M | apache-2.0 | ZaandaTeika | 1 | 721 |
| itlwas/Qwen2.5-Math-7B-Instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | itlwas | 0 | 722 |
| HuggingFaceH4/Qwen2.5-Math-7B-Instruct-PRM-0.2 | finetune | — | 7.07B | M | — | HuggingFaceH4 | 0 | 654 |
| Mosfets/Qwen2.5-7B-Instruct-MathCoder-GGUF | merge | GGUF | — | — | — | Mosfets | 0 | 634 |
| Lawnakk/BBALAW1.0 | merge | — | 4.35B | M | — | Lawnakk | 0 | 630 |
| DuoNeural/Qwen2.5-Math-NeuralMath-7B | adapter | GGUF | 7.62B | M | apache-2.0 | DuoNeural | 0 | 572 |
| DongfuJiang/prm_qwen25_math_gsm_2k_with_full_sol_mix_ref_redistribution_hf | finetune | — | 7.62B | M | apache-2.0 | DongfuJiang | 0 | 567 |
| flyingbugs/OpenR1-Qwen-7B-SFT | finetune | — | 7.62B | M | — | flyingbugs | 1 | 474 |
Hosted endpoints observed for Qwen2.5-Math-7B-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 | Qwen/Qwen2.5-Math-7B-Instruct | — | — | — | — | — | — | — | 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 21:39 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:11 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.
