Up to date
HuggingFaceH4 MIT out 3 years ago
7.24B parameters, text in, text out, 32K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #488 in LLMs · score 76.9 method v1 Sources 3 Downloads 30d 5k · #545 of 619 tracked
Changed this week: 3 new quantized builds on the Hub (13 in all).2026-10-24
Picked by rule from 5 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 2 picks above, plus these 3. 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; 13 quantized builds for local use; a big base for builders (63 fine-tunes, adapters and merges).
Top Spaces running it: baconnier/prompt-plus-plus ♥ 246 · NiansuhAI/HFLLMs ♥ 30 · ysharma/Zephyr-Playground ♥ 17 · radinhas/hf-llm-api ♥ 14
Weights only, an estimate: leave room for the KV cache. Method
About 4.3 GB at INT4, 8.7 GB at FP8, 17 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 (2023-10-09), amber triangles are 4 dated articles (each links to it), grey ticks are repo commits.
76 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.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported Logits/chosen | -2.74 | — | model author | 2026-10-07 |
| author-reported Logits/rejected | -2.72 | — | model author | 2026-10-07 |
| author-reported Logps/chosen | -297.1 | — | model author | 2026-10-07 |
| author-reported Logps/rejected | -327.43 | — | model author | 2026-10-07 |
| author-reported Loss | 0.46 | — | model author | 2026-10-07 |
| author-reported Rewards/accuracies | 0.78 | — | model author | 2026-10-07 |
| author-reported Rewards/chosen | -0.51 | — | model author | 2026-10-07 |
| author-reported Rewards/margins | 1.37 | — | model author | 2026-10-07 |
| author-reported Rewards/rejected | -1.88 | — | model author | 2026-10-07 |
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 | MistralForCausalLM |
| Model type | mistral |
| Dtype | bfloat16 |
| Params by dtype | BF16: 7,241,732,096 |
| Size band | M |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Languages | en |
| Papers | arXiv:2305.18290, arXiv:2310.16944, arXiv:2305.14233, arXiv:2310.01377 |
| License | MIT (mit) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · MistralForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · MistralForCausalLM · 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 zephyr-7b-alpha: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 17 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 8.7 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.3 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Base model: mistralai/Mistral-7B-v0.1
quantized: 13merge: 4finetune: 14adapter: 45
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| TheBloke/zephyr-7B-alpha-AWQ | quantized | AWQ | 7.24B | M | mit | TheBloke | 18 | 136.2k |
| TheBloke/zephyr-7B-alpha-GPTQ | quantized | GPTQ | 7.24B | M | mit | TheBloke | 26 | 121.3k |
| TheBloke/zephyr-7B-alpha-GGUF | quantized | GGUF | — | — | mit | TheBloke | 138 | 51.6k |
| sethuiyer/SynthIQ-7b | merge | — | 7.24B | M | llama2 | sethuiyer | 3 | 21.6k |
| mradermacher/zephyr-7b-alpha-GGUF | quantized | GGUF | — | — | mit | mradermacher | 0 | 9.2k |
| MaziyarPanahi/zephyr-7b-alpha-GGUF | quantized | GGUF | — | — | apache-2.0 | MaziyarPanahi | 0 | 7.5k |
| QuantFactory/zephyr-7b-alpha-GGUF | quantized | GGUF | — | — | mit | QuantFactory | 0 | 7k |
| second-state/Zephyr-7B-Alpha-GGUF | quantized | GGUF | — | — | mit | second-state | 0 | 6.4k |
| tensorblock/zephyr-7b-alpha-GGUF | quantized | GGUF | — | — | mit | tensorblock | 0 | 2k |
| monsterapi/zephyr-7b-alpha_metamathqa | finetune | — | 7.24B | M | apache-2.0 | monsterapi | 4 | 849 |
| Sakalti/Magro-7b-v1.1 | merge | — | 7.24B | M | mit | Sakalti | 0 | 823 |
| linxule/zephyr-7b-alpha.Q4_K_M | quantized | GGUF | — | — | mit | linxule | 0 | 609 |
| monsterapi/zephyr_7b_norobots | adapter | — | — | — | apache-2.0 | monsterapi | 2 | 304 |
| liodon-ai/zephyr-7b-alpha-imatrix-GGUF | quantized | imatrix | — | — | other | liodon-ai | 1 | 289 |
| liodon-ai/zephyr-7b-alpha-ONNX | quantized | — | — | — | other | liodon-ai | 0 | 166 |
| eek/zephyr-7b-sft-lora | adapter | — | — | — | mit | eek | 0 | 144 |
| monsterapi/zephyr_7b_WizardLMEvolInstruct70k | adapter | — | — | — | apache-2.0 | monsterapi | 0 | 135 |
| mzhaoshuai/zephyr-7b-alpha-conf-sft | quantized | — | — | — | apache-2.0 | mzhaoshuai | 0 | 127 |
| koussayyyy/zephyr_automotive_testcase_model | adapter | — | — | — | mit | koussayyyy | 2 | 23 |
| BadreddineHug/zyph_OCR | adapter | — | — | — | — | BadreddineHug | 0 | 120 |
| theBodhiTree/Zephyr_7B_AlphaBeta_TaskArithmetic | finetune | — | 7.11B | M | — | theBodhiTree | 1 | 69 |
| shaktiman404/ZephyrChat-Mistral-7B | merge | — | 7.24B | M | — | shaktiman404 | 1 | 64 |
| tourist800/ORKG-zephyr-7b-alpha-finetune | adapter | — | — | — | — | tourist800 | 0 | 106 |
| Teera/Zephyr-7B-Relation-Extraction-fully | adapter | — | — | — | — | Teera | 0 | 85 |
| Ghali20/test_1 | adapter | — | — | — | — | Ghali20 | 1 | 33 |
| RichardMJ/my-Zephyr1 | adapter | — | — | — | — | RichardMJ | 0 | 77 |
| masonanalytics/HuggingFaceH4-zephyr-7b-alpha-FullChat-LORA-512 | adapter | — | — | — | — | masonanalytics | 0 | 70 |
| Fermat111/Prueba9927Zephyr | adapter | — | — | — | — | Fermat111 | 0 | 68 |
| xiaorui638/mistral_merged8_ties | merge | — | 7.24B | M | — | xiaorui638 | 0 | 63 |
| HadrousPrime/SDS_Fine_Tune_Qlora | adapter | — | — | — | — | HadrousPrime | 0 | 61 |
| masonanalytics/PEFT-Zephyr-7B-Alpha | adapter | — | — | — | — | masonanalytics | 0 | 57 |
| masonanalytics/HuggingFaceH4-zephyr-7b-alpha-20231105-1024Tokens-lora | adapter | — | — | — | — | masonanalytics | 0 | 42 |
| AswanthCManoj/azma-zephyre-lora-adapter | adapter | — | — | — | — | AswanthCManoj | 0 | 40 |
| BadreddineHug/Zephyr_finetuned | adapter | — | — | — | — | BadreddineHug | 0 | 40 |
| masonanalytics/PEFT-Zephyr-7B-Alpha_8bit | adapter | — | — | — | — | masonanalytics | 0 | 38 |
| llm-wizard/sft_zephyr | adapter | — | — | — | mit | llm-wizard | 0 | 37 |
| BadreddineHug/zeyphr_5Ms | adapter | — | — | — | — | BadreddineHug | 0 | 35 |
| iagoalves/zephyr_7b_enem_apostilas_1500steps_constant_zzzzzzz | adapter | — | — | — | mit | iagoalves | 0 | 35 |
| BadreddineHug/zyph_5M_167_rows | adapter | — | — | — | — | BadreddineHug | 0 | 34 |
| BadreddineHug/zephyr_test_1 | adapter | — | — | — | — | BadreddineHug | 0 | 34 |
| Date | Commit | SHA |
|---|---|---|
| 2024-10-16 | Update README.md | 014792b |
| 2023-11-21 | Update README.md | 2ce2d02 |
| 2023-10-26 | Add base_model tag | f28e1c0 |
| 2023-10-17 | Update README.md | 5d2b362 |
| 2023-10-16 | Upload colab-demo.ipynb | 2140056 |
| 2023-10-16 | Shard checkpoints in 2GB chunks to run on Colab | fd8e906 |
| 2023-10-14 | Delete dialogue_template.json | 080e169 |
| 2023-10-13 | Upload thumbnail.png | 3d9d09e |
| 2023-10-13 | Update README.md | 4d16a3f |
| 2023-10-13 | Change license to MIT 🪁 | 3e8baad |
| 2023-10-11 | Update README.md | ad12cca |
| 2023-10-11 | Add chat template to README | aa57c05 |
| 2023-10-11 | Add chat template (#2) | 9894b57 |
| 2023-10-10 | Update README.md | 19ac753 |
| 2023-10-10 | Update README.md | 2cd2cd1 |
| 2023-10-10 | Upload thumbnail.png | 366696a |
| 2023-10-10 | Update README.md | 4372f6c |
| 2023-10-10 | Fix EOS token | c0b9e47 |
| 2023-10-09 | Update config.json | 284563e |
| 2023-10-09 | Add HuggingFaceH4/mistral-7b-dpo-v0.4 checkpoint | ed8ea81 |
| 2023-10-09 | initial commit | 7e6e85d |
Hosted endpoints observed for zephyr-7b-alpha, 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 | HuggingFaceH4/zephyr-7b-alpha | — | — | — | — | — | — | — | 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.
| Title | Source | Published | First seen | Signals |
|---|---|---|---|---|
| StableLM Zephyr 3B | stability.ai | 2023-12-07 | 2026-10-07 | hn |
| ZSWatch – Open-source Zephyr-based smartwatch | github.com | 2023-01-14 | 2026-10-07 | hn |
| Zero-install embedded training with Zephyr using Kasm and Docker | blog.golioth.io | 2022-07-14 | 2026-10-07 | hn |
| Zephyr: A scalable Real-Time operating system for IoT embedded devices | zephyrproject.org | 2019-11-29 | 2026-10-07 | hn |
| Meet Linux’s little brother: Zephyr, a tiny open-source IoT RTOS | linuxgizmos.com | 2016-02-18 | 2026-10-07 | hn |
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 21:40 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:12 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.
