Up to date
1.42B parameters, text in, text out, 2K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #590 in LLMs · score 74.3 method v1 Sources 3 Downloads 30d 20.8k · #453 of 619 tracked
Picked by rule from 2 tracked pieces: the lab's own post, the most shared, a field report, the latest.
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; 8 quantized builds for local use.
Top Spaces running it: LanguageBind/MoE-LLaVA ♥ 154 · Yeyito/llm_contamination_detector ♥ 27 · mPLUG/TinyChart-3B ♥ 17 · showlab/Show-o ♥ 17
Weights only, an estimate: leave room for the KV cache. Method
About 0.9 GB at INT4, 1.7 GB at FP8, 3.4 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 (2023-09-10), amber triangles are 2 dated articles (each links to it), grey ticks are repo commits.
22 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 | PhiForCausalLM |
| Model type | phi |
| Dtype | float32 |
| Params by dtype | F16: 1,418,270,720 |
| Size band | S |
| MoE | no |
| Context length | 2,048 (config.json max_position_embeddings) |
| Languages | en |
| Papers | arXiv:2306.11644 |
| License | MIT (mit) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · PhiForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · PhiForCausalLM · 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 phi-1: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 3.4 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 1.7 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.9 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 8merge: 1finetune: 3adapter: 10
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| mradermacher/phi-1-i1-GGUF | quantized | GGUF | — | — | mit | mradermacher | 0 | 3.6k |
| mradermacher/phi-1-GGUF | quantized | GGUF | — | — | mit | mradermacher | 1 | 2.3k |
| professorf/phi-1-gguf | quantized | gguf | — | — | mit | professorf | 1 | 1.7k |
| itlwas/phi-1-Q4_K_M-GGUF | quantized | GGUF | — | — | mit | itlwas | 0 | 353 |
| leliuga/phi-1-bnb-4bit | quantized | -bnb | 1.46B | S | mit | leliuga | 0 | 280 |
| Phudish/Test_Tam_2 | merge | — | 1.53B | S | — | Phudish | 0 | 273 |
| Merhabalad/phi-1-Q3_K_M-GGUF | quantized | GGUF | — | — | mit | Merhabalad | 0 | 270 |
| MatteoKhan/phi-1-1.5-merged | finetune | — | 1.42B | S | mit | MatteoKhan | 0 | 161 |
| Mayank-sharma108/phi-1-Q4_K_M-GGUF | quantized | GGUF | — | — | mit | Mayank-sharma108 | 0 | 135 |
| Judiht/ludwigfinetuned_phi-1_testcase_generator_20k | adapter | — | — | — | — | Judiht | 0 | 116 |
| TheOSExplorer/phi-1-Q2_K-GGUF | quantized | Q2_K | — | — | mit | TheOSExplorer | 0 | 114 |
| kyryl-opens-ml/phi-1-duckdb-text-to-sql | adapter | — | — | — | mit | kyryl-opens-ml | 0 | 40 |
| rk68/phi-1-finetuned-aqua-rat-2k | adapter | — | — | — | mit | rk68 | 0 | 39 |
| AKASH2393/phi1-lora-bilingual-ft | adapter | — | — | — | — | AKASH2393 | 0 | 39 |
| rk68/phi-1-finetuned-aqua-rat-teacher-2k | adapter | — | — | — | mit | rk68 | 0 | 36 |
| Judiht/finetuned_phi-1_testcase_generator_20k_v2 | adapter | — | — | — | — | Judiht | 0 | 21 |
| Judiht/finetuned_phi-1_testcase_generator_20k-lora | adapter | — | — | — | — | Judiht | 0 | 19 |
| Judiht/finetuned_phi-1_testcase_generator-lora | adapter | — | — | — | — | Judiht | 0 | 15 |
| Judiht/ludwigfinetuned_phi-1_testcase_generator_1k | adapter | — | — | — | — | Judiht | 0 | 15 |
| raghu60/phi1-lora-finetuned | adapter | — | — | — | — | raghu60 | 0 | 12 |
| IyedLahiani/HTML_CSS_CodeGenerator | finetune | — | — | — | — | IyedLahiani | 0 | 0 |
| iravikr/phi1-numpy-pandas-qlora | finetune | — | — | — | apache-2.0 | iravikr | 0 | 0 |
| Date | Commit | SHA |
|---|---|---|
| 2025-11-24 | Update README.md (#15) | d4c0adc |
| 2025-11-24 | Upload data_summary_card.md (#14) | c542f61 |
| 2024-04-29 | fix(config): Removes auto_map since it is not used anymore. | b9ac0e6 |
| 2024-04-23 | Update README.md | cb3004d |
| 2024-04-23 | Delete modeling_phi.py | 8399ff6 |
| 2024-04-23 | Delete configuration_phi.py | 8c83568 |
| 2024-04-23 | Update README.md | 92f2355 |
| 2024-04-17 | Delete pytorch_model.bin | 0c849e4 |
| 2024-04-17 | Adding `safetensors` variant of this model (#9) | b336aa7 |
| 2024-02-06 | Update LICENSE | a1a5931 |
| 2024-01-24 | Update README.md | ac0061a |
| 2024-01-22 | Update README.md | ffff260 |
| 2024-01-22 | Update README.md | 2aef707 |
| 2024-01-22 | Update README.md | 64cc1d8 |
| 2024-01-22 | Update config.json | 0d846e0 |
| 2024-01-16 | Update modeling_phi.py | 944a013 |
| 2024-01-16 | Update README.md | fb32aac |
| 2024-01-16 | Update README.md | 07d9363 |
| 2024-01-12 | Update modeling_phi.py | 03b9f69 |
| 2024-01-12 | Update modeling_phi.py | 54bed1a |
| 2024-01-11 | Update modeling_phi.py | 50bb267 |
| 2024-01-10 | Upload modeling_phi.py | 2cfa65f |
| 2024-01-09 | Delete Research License.docx | 957a783 |
| 2024-01-09 | Upload 5 files | 1cb0668 |
| 2024-01-08 | Update config.json | 3e53f58 |
| 2024-01-08 | Update modeling_phi.py | e575241 |
| 2024-01-05 | Update modeling_phi.py | 4e6ed66 |
| 2024-01-05 | Update configuration_phi.py | fbf395a |
| 2024-01-05 | fix(root): Fixes relative paths. | b908838 |
| 2024-01-05 | chore(root): Updates files to internal transformers implementation. | 8a2c68b |
Hosted endpoints observed for phi-1, 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 | microsoft/phi-1 | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency.
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 21:42 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:14 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.
