Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Models › Qwen1.5 family

Qwen1.5-4B-Chat

Hugging Face ↗ Follow changes

Qwen Research / non-commercial (tongyi-qianwen-research) out 3 years ago

3.95B parameters, text in, text out. 1 provider serves it.

Verified 2026-10-25 Rank #635 in LLMs · score 73.3 method v1 Sources 3 Downloads 30d 16.1k · #473 of 619 tracked

Can I run it?2.4GB at INT4Estimatefits a 16 GB cardmethod
Can I use it?Research / non-commercial (tongyi-qianwen-research)Non-commercial onlylicence file
Do people use it?16.1kdownloads, 30 days#473 of 619 we track · below the median+2 likes, downloads −7% this week52 likesHugging Face

Worth reading

Picked by rule from 7 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Latest · 3 years ago

Qwen1.5-Moe: Matching 7B Model Performance with 1/3 Activated Parameters

Qwen1.5-MoE-A2.7B matches 7B model performance with only 2.7 billion activated parameters using a 64-expert configuration.

  • The model partitions a single FFN into segments to generate 64 experts without increasing the total parameter count.
  • Training from scratch is inefficient, so the team repurposes the existing Qwen-1.8B model to initialize Qwen1.5-MoE-A2.7B.
qwenlm.github.io 2024-03-28 104 HN pts · 10 comments 1.5k words about the Qwen1.5 family family

7 best of 7 tracked

The 1 pick above, plus these 6. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

  1. ArticleQwen1.5-110Bqwenlm.github.io · 2024-04-26 · official
  2. ArticleQwen1.5-32B: Fitting the Capstone of the Qwen1.5 Language Model Seriesqwenlm.github.io · 2024-04-05 · official
  3. ArticleQwen1.5qwenlm.github.io · 2024-02-05 · 5 HN pts · official
  4. DiscussionQwen1.5-110B Previewtwitter.com · 2024-04-20
  5. ArticleComplete tutorial: Run Qwen LLM across your local machines without Python/C++secondstate.io · 2024-02-22
  6. ArticleRun Alicloud's Qwen LLM with a 2MB cross GPU/CPU rust appsecondstate.io · 2024-02-28

Where to run it

Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.

#Cheapest live rowIn /1MOut /1MContextQuantSeen
1 Featherless — — — — Oct 25

All 1 provider →

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.

Served by 1 API provider; 17 quantized builds for local use.

  • As a service1Featherless
  • On own hardware17quantized builds
  • Built on273 fine-tunes · 22 adapters · 2 merges
  • In apps0Hugging Face Spaces

Will it fit?

Weights only, an estimate: leave room for the KV cache. Method

Fits a 16 GB card at INT4

About 2.4 GB at INT4, 4.7 GB at FP8, 9.5 GB at BF16: weights only, so leave room for the KV cache.

16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.4 GB of weights (estimate; KV cache and activations extra)INT42.4 GBFP8: about 4.7 GB of weights (estimate; KV cache and activations extra)FP84.7 GBBF16: about 9.5 GB of weights (estimate; KV cache and activations extra)BF169.5 GB
16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.4 GB of weights (estimate; KV cache and activations extra)INT42.4 GBFP8: about 4.7 GB of weights (estimate; KV cache and activations extra)FP84.7 GBBF16: about 9.5 GB of weights (estimate; KV cache and activations extra)BF169.5 GB

Estimate params × bytes per param × 1.2 · how we estimate

The story so far

Weights, every dated article, and repo commits on one rail.

2 dated articles and 0 repo commits so far

Teal is the weights (2024-01-30), amber triangles are 2 dated articles (each links to it).

FebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctQwen1.5-Moe: Matching 7B Model Performance with 1/3 Activated Parameters · 2024-03-28Qwen1.5 · 2024-02-05Weights 2024-01-30 · 2024-01-30Weights 2024-01-30now · 1 provider
FebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctQwen1.5-Moe: Matching 7B Model Performance with 1/3 Activated Parameters · 2024-03-28Qwen1.5 · 2024-02-05Weights 2024-01-30 · 2024-01-30Weights 2024-01-30now · 1 provider

What people built on it

44 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

44 derivatives on the Hub, most of them adapters

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

17 quantized (Hugging Face model tree)3 fine-tunes (Hugging Face model tree)2 merges (Hugging Face model tree)22 adapters (Hugging Face model tree)17 quantized3 fine-tunes2 merges22 adapters
17 quantized (Hugging Face model tree)3 fine-tunes (Hugging Face model tree)2 merges (Hugging Face model tree)22 adapters (Hugging Face model tree)17 quantized3 fine-tunes2 merges22 adapters
GGUF second-state/Qwen1.5-4B-Chat-GGUF for llama.cpp, Ollama, LM Studio 9.7k downloads

All variants →

Reference

Everything else, folded: the full tables, every source.

Use-case evidence

2 use cases

What it is good at 2 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.

Facts

15 fields

Listed in LLMs

Facts

Pipeline tagtext-generation
Librarytransformers
ArchitecturesQwen2ForCausalLM
Model typeqwen2
Params by dtypeBF16: 3,950,369,280
Size bandM
MoEno
PapersarXiv:2309.16609
LicenseResearch / non-commercial (tongyi-qianwen-research) (other)
Hub snapshot2026-10-25

Hardware fit

3 estimates

Hardware fit ESTIMATE

Weights-only memory for Qwen1.5-4B-Chat: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)9.5 GB · Single GPU (<24 GB)
FP8 (1 byte/param)4.7 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)2.4 GB · Edge (<4 GB)

Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.

All variants

40 variants

Lineage

Original model Family: Qwen/Qwen1.5 44 derivatives on the Hub

quantized: 17adapter: 22merge: 2finetune: 3

Top variants (40 shown, ranked by downloads)

Relation
Quant
40 rows
Variant Relation Quant Params Band License Author Likes Downloads
second-state/Qwen1.5-4B-Chat-GGUF quantized GGUF — — other second-state 0 9.7k
mradermacher/Qwen1.5-4B-Chat-i1-GGUF quantized GGUF — — other mradermacher 0 3.1k
CHE-72/Qwen1.5-4B-Chat-Q2_K-GGUF quantized Q2_K — — other CHE-72 0 2.5k
tensorblock/Qwen1.5-4B-Chat-GGUF quantized GGUF — — other tensorblock 0 2.2k
mradermacher/Qwen1.5-4B-Chat-GGUF quantized GGUF — — other mradermacher 0 2.1k
CHE-72/Qwen1.5-4B-Chat-Q6_K-GGUF quantized Q6_K — — other CHE-72 0 1.1k
CHE-72/Qwen1.5-4B-Chat-Q8_0-GGUF quantized Q8_0 — — other CHE-72 0 988
CHE-72/Qwen1.5-4B-Chat-Q5_K_M-GGUF quantized GGUF — — other CHE-72 0 968
svjack/Genshin_Impact_Qwen_1_5_Plot_Engine_Step_Half_lora_small adapter GGUF — — other svjack 0 848
CHE-72/Qwen1.5-4B-Chat-Q4_0-GGUF quantized Q4_0 — — other CHE-72 0 846
CHE-72/Qwen1.5-4B-Chat-Q4_K_M-GGUF quantized GGUF — — other CHE-72 0 811
GenTrendGPT/OS-Test-Mark-GEN-I merge — 4.10B M — GenTrendGPT 0 792
CHE-72/Qwen1.5-4B-Chat-Q5_K_S-GGUF quantized GGUF — — other CHE-72 0 782
CHE-72/Qwen1.5-4B-Chat-Q3_K_L-GGUF quantized GGUF — — other CHE-72 0 715
CHE-72/Qwen1.5-4B-Chat-Q4_K_S-GGUF quantized GGUF — — other CHE-72 0 709
CHE-72/Qwen1.5-4B-Chat-Q3_K_M-GGUF quantized GGUF — — other CHE-72 0 675
CHE-72/Qwen1.5-4B-Chat-Q3_K_S-GGUF quantized GGUF — — other CHE-72 0 608
MadeAgents/Hammer-4b finetune — 3.95B M other MadeAgents 3 457
itlwas/Qwen1.5-4B-Chat-Q4_K_M-GGUF quantized GGUF — — other itlwas 0 452
GenTrendGPT/OS-Test-Mark-GEN merge — 3.80B M — GenTrendGPT 0 402
clepelaars/task-7-Qwen-Qwen1.5-4B-Chat adapter — 4.00B M — clepelaars 0 373
Atmanstr/qwen-financebot-finetuned adapter — — — — Atmanstr 1 273
CHE-72/Qwen1.5-4B-Chat-Q5_0-GGUF quantized Q5_0 — — other CHE-72 0 203
langgptai/Qwen-sft-la-v0.1 adapter — — — other langgptai 0 145
svjack/Genshin_Impact_Qwen_1_5_Plot_Engine_Step_Json_Short_lora_small adapter — — — other svjack 0 112
mikethoue/task-16-Qwen-Qwen1.5-4B-Chat adapter — — — — mikethoue 0 106
dir00/task-13-Qwen-Qwen1.5-4B-Chat adapter — — — — dir00 0 53
jfranklin-foundry/task-3-Qwen-Qwen1.5-4B-Chat adapter — — — — jfranklin-foundry 0 48
gerbyk/task-13-Qwen-Qwen1.5-4B-Chat adapter — — — — gerbyk 0 44
langgptai/Qwen-sft-ls-v0.1 adapter — — — other langgptai 0 36
svjack/Genshin_Impact_Qwen_1_5_Plot_Engine_Step_Json_lora_small adapter — — — other svjack 0 35
santyzenith/Qwen-1-5-4B-UDA-Rules-SFT adapter — — — other santyzenith 0 33
ycheng1024/task-17-Qwen-Qwen1.5-4B-Chat adapter — — — — ycheng1024 0 33
langgptai/Qwen-las-v0.1 adapter — — — other langgptai 0 29
ZeroZYbgp/pocket_nahida-4b-lora adapter — — — — ZeroZYbgp 0 28
313505011chiayu/sobel-edge-gpu adapter — — — — 313505011chiayu 0 26
313505011chiayu/HW4_model_weight adapter — — — — 313505011chiayu 0 22
gerbyk/task-16-Qwen-Qwen1.5-4B-Chat adapter — — — — gerbyk 0 20
dir00/task-16-Qwen-Qwen1.5-4B-Chat adapter — — — — dir00 0 20
mikekubi/task-13-Qwen-Qwen1.5-4B-Chat adapter — — — — mikekubi 0 17

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for Qwen1.5-4B-Chat, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.

ProviderModel refContextQuant$/M in$/M outtok/sTTFTToolsStatusObserved
Featherless
hf
 Qwen/Qwen1.5-4B-Chat — — — — — — — live 2026-10-25

Articles and coverage

7 pieces

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.

Official2

Qwen1.5-Moe: Matching 7B Model Performance with 1/3 Activated Parameters
qwenlm.github.io2024-03-28officialauthorityHN front pageabout the Qwen1.5 familyHN 104 pts · 10 comments1,495 words
Qwen1.5-MoE-A2.7B matches 7B model performance with only 2.7 billion activated parameters using a 64-expert configuration. The model incorporates 4 shared experts alongside 60 routing experts and requires source installation via git for Hugging Face support. It achieves a 75% decrease in training expenses and accelerates inference speed by a factor of 1.74 compared to Qwen1.5-7B.written, reviewed, cited
  • The model partitions a single FFN into segments to generate 64 experts without increasing the total parameter count.
  • Training from scratch is inefficient, so the team repurposes the existing Qwen-1.8B model to initialize Qwen1.5-MoE-A2.7B.
  • The Qwen1.5-MoE-A2.7B-Chat-GPTQ-Int4 quantized variant is available, but AWQ support is currently not supported.
Qwen1.5
qwenlm.github.io2024-02-05officialauthorityabout the Qwen1.5 familyHN 5 pts · 2 comments3,793 words
Qwen1.5 is a new open-source model series released before the Chinese New Year, offering base and chat models in six sizes ranging from 0.5B to 110B parameters. The Qwen1.5-72B model outperforms Llama2-70B across all benchmarks, and the series features improved multilingual and human-alignment capabilities. All models now uniformly support a context length of up to 32768 tokens, and the Qwen1.5-72B-Chat model is recommended for initial testing.written, reviewed, cited
All coverage tracked for this model (7) — append-only, nothing is ever removed
TitleSourcePublishedFirst seenSignals
Qwen1.5-110Bqwenlm.github.io official2024-04-262026-10-09hn
Qwen1.5-110B Previewtwitter.com2024-04-202026-10-09hn
Qwen1.5-32B: Fitting the Capstone of the Qwen1.5 Language Model Seriesqwenlm.github.io official2024-04-052026-10-09hn
Qwen1.5-Moe: Matching 7B Model Performance with 1/3 Activated Parametersqwenlm.github.io official2024-03-282026-10-09hn
Run Alicloud's Qwen LLM with a 2MB cross GPU/CPU rust appsecondstate.io2024-02-282026-10-09hn
Complete tutorial: Run Qwen LLM across your local machines without Python/C++secondstate.io2024-02-222026-10-09hn
Qwen1.5qwenlm.github.io official2024-02-052026-10-09hn

Papers

1 paper

Papers 1

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).

Similar models

3 models

Similar models

  • Qwen3-4B · 4.02B
    same size band (M), same task (text-generation), permissive license
  • Qwen3-4B-Instruct-2507 · 4.02B
    same size band (M), same task (text-generation), permissive license
  • Qwen3-4B-Thinking-2507 · 4.02B
    same size band (M), same task (text-generation), more providers, permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-09 02:04 UTC · Last verified 2026-10-25

Follow Qwen1.5-4B-Chat

Watch this model

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.