Up to date
google Gemma Terms of Use out 20 months ago
1000M parameters, text in, text out, 32K tokens of context. 1 provider serves it.
Verified 2026-10-26 07:27 UTC Rank #73 in LLMs · score 103.9 method v1 Sources 3 Downloads 30d 839.3k · #125 of 619 tracked
Changed this week: 5 new quantized builds on the Hub (486 in all).2026-10-24 · 4 new quantized builds on the Hub (485 in all).2026-10-24
Picked by rule from 35 tracked pieces: the lab's own post, the most shared, a field report, the latest.
MetalChat is a Metal-accelerated C++ framework and command line interpreter for inference of Meta Llama, Google Gemma, and Liquid models.
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 26 |
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 (486 quantized builds); a big base for builders (794 fine-tunes, adapters and merges).
Top Spaces running it: embedl/hfviewer ♥ 181 · hadadrjt/ai ♥ 86 · garage-lab/MCP_HTML2JSON ♥ 33 · Monster/gemma-3-1b-it ♥ 27
Weights only, an estimate: leave room for the KV cache. Method
About 0.6 GB at INT4, 1.2 GB at FP8, 2.4 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Third-party measurements only; author-reported numbers stay in Reference, marked as such.
7.17. Each tick is a model on this site with a measured BFCL row; author-reported numbers stay in Reference. Percent (overall accuracy). Prompt
Source · leaderboard · 2026-10-26
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2025-03-10), amber triangles are 8 dated articles (each links to it), grey ticks are repo commits.
1,280 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 |
|---|---|---|---|---|
| measured BFCL · Call tools and APIs | 7.17 percent (overall accuracy) | Prompt5 sub-scores
| leaderboard | 2026-10-26 |
| author-reported ARC Challenge · Chat and follow instructions | 38.4 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported BBH · Reason through hard problems | 28.4 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 63.2 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 73.8 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 48.9 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 39.8 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 9.48 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Commonsense QA sets · Chat and follow instructions | 73 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported DROP | 42.4 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported HellaSwag · Chat and follow instructions | 62.3 | Gemma 3 PT 1B | model author | 2026-10-07 |
| author-reported Winogrande · Chat and follow instructions | 58.2 | Gemma 3 PT 1B | model author | 2026-10-07 |
Gemma 3 1B IT is a multimodal instruction-tuned model with 1000M parameters.[1][2] The model handles text and image inputs to generate text outputs.[3] It is released under the Gemma Terms of Use and supports a 32K token input context.[4][5]
Generated 2026-10-07 15:00 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 |
|---|---|
| Library | transformers |
| Architectures | Gemma3ForCausalLM |
| Model type | gemma3_text |
| Dtype | bfloat16 |
| Params by dtype | BF16: 999,885,952 |
| Size band | S |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Gated | manual |
| Papers | arXiv:1905.07830, arXiv:1905.10044, arXiv:1911.11641, arXiv:1904.09728, arXiv:1705.03551, arXiv:1911.01547, arXiv:1907.10641, arXiv:1903.00161, arXiv:2009.03300, arXiv:2304.06364, arXiv:2103.03874, arXiv:2110.14168, arXiv:2311.12022, arXiv:2108.07732, arXiv:2107.03374, arXiv:2210.03057, arXiv:2106.03193, arXiv:1910.11856, arXiv:2502.12404, arXiv:2502.21228, arXiv:2404.16816, arXiv:2104.12756, arXiv:2311.16502, arXiv:2203.10244, arXiv:2404.12390, arXiv:1810.12440, arXiv:1908.02660, arXiv:2312.11805 |
| License | Gemma Terms of Use (gemma) |
| Hub snapshot | 2026-10-26 |
vLLM: supported · Gemma3ForCausalLM · 4b1c04e76fe9ac3ca8b7e541dd423e7d5f38dc60SGLang: supported · Gemma3ForCausalLM · 60d941293e3af27a12c6eb64e2c3e2e1a17f100f
vLLM registry (4b1c04e76fe9ac3ca8b7e541dd423e7d5f38dc60) · SGLang registry (60d941293e3af27a12c6eb64e2c3e2e1a17f100f) · 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 gemma-3-1b-it: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 2.4 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 1.2 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.6 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Base model: google/gemma-3-1b-pt
quantized: 486finetune: 572merge: 10adapter: 212
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Andycurrent/Gemma-3-1B-it-GLM-4.7-Flash-Heretic-Uncensored-Thinking_GGUF | quantized | GGUF | — | — | gemma | Andycurrent | 165 | 7.9M |
| MaziyarPanahi/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | — | MaziyarPanahi | 13 | 3.5M |
| unsloth/gemma-3-1b-it | finetune | — | 1000M | S | gemma | unsloth | 22 | 1.6M |
| unsloth/gemma-3-1b-it-unsloth-bnb-4bit | quantized | -bnb | 1.02B | S | gemma | unsloth | 12 | 1.2M |
| ggml-org/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | ggml-org | 54 | 1.1M |
| unsloth/gemma-3-1b-it-bnb-4bit | quantized | -bnb | 1.02B | S | gemma | unsloth | 10 | 1.1M |
| farbodtavakkoli/OTel-LLM-1B-IT | finetune | — | 1000M | S | apache-2.0 | farbodtavakkoli | 1 | 908.9k |
| unsloth/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | unsloth | 95 | 853.9k |
| mlx-community/gemma-3-1b-it-qat-4bit | quantized | -4bit | 1.30B | S | gemma | mlx-community | 6 | 746.5k |
| lmstudio-community/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | lmstudio-community | 9 | 358k |
| lmstudio-community/gemma-3-1B-it-qat-GGUF | quantized | GGUF | — | — | gemma | lmstudio-community | 19 | 226.4k |
| bartowski/google_gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | bartowski | 20 | 133.7k |
| mlx-community/gemma-3-1b-it-4bit | quantized | -4bit | 1.30B | S | gemma | mlx-community | 6 | 94.8k |
| RedHatAI/gemma-3-1b-it-quantized.w8a8 | quantized | -quant | 1.30B | S | gemma | RedHatAI | 2 | 89.9k |
| tetf/gemma-3-1b-it-qat-q4_0-GGUF | quantized | q4_0 | — | — | gemma | tetf | 0 | 68k |
| gaunernst/gemma-3-1b-it-int4-awq | quantized | int4 | 1000M | S | gemma | gaunernst | 2 | 55.6k |
| google/gemma-3-1b-it-qat-q4_0-gguf | quantized | q4_0 | — | — | gemma | 174 | 35.7k | |
| mlc-ai/gemma-3-1b-it-q4f16_1-MLC | quantized | — | — | — | — | mlc-ai | 0 | 26.1k |
| gaunernst/gemma-3-1b-it-qat-compressed-tensors | quantized | compressed-tensors | 1.02B | S | gemma | gaunernst | 0 | 25.4k |
| onnx-community/gemma-3-1b-it-ONNX | quantized | — | — | — | gemma | onnx-community | 27 | 21.3k |
| google/gemma-3-1b-it-qat-int4-unquantized | finetune | int4 | 1000M | S | gemma | 14 | 21.3k | |
| DavidAU/Gemma-3-1b-it-MAX-NEO-Imatrix-GGUF | quantized | Imatrix | — | — | apache-2.0 | DavidAU | 6 | 20.3k |
| google/gemma-3-1b-it-qat-q4_0-unquantized | finetune | q4_0 | 1000M | S | gemma | 10 | 17.5k | |
| second-state/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | second-state | 0 | 17.3k |
| reaperdoesntknow/Qemma-Q14B | merge | — | 1.01B | S | osl-3.0 | reaperdoesntknow | 3 | 16.5k |
| reaperdoesntknow/Qemma-Q1.7B | merge | — | 1000M | S | osl-3.0 | reaperdoesntknow | 0 | 15.8k |
| onnx-community/gemma-3-1b-it-ONNX-GQA | quantized | — | — | — | — | onnx-community | 15 | 13.4k |
| DavidAU/gemma-3-1b-it-heretic-extreme-uncensored-abliterated | finetune | — | 1000M | S | — | DavidAU | 62 | 10.2k |
| mlx-community/gemma-3-1b-it-4bit-DWQ | quantized | -4bit | 1.30B | S | gemma | mlx-community | 0 | 13k |
| mlx-community/gemma-3-1b-it-8bit | quantized | -8bit | 1.30B | S | gemma | mlx-community | 5 | 12.3k |
| matrixportalx/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | matrixportalx | 0 | 12.5k |
| gguf-org/gemma-3-1b-it-gguf | quantized | gguf | — | — | gemma | gguf-org | 1 | 11.4k |
| brittlewis12/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | brittlewis12 | 2 | 11.3k |
| RedHatAI/gemma-3-1b-it-quantized.w4a16 | quantized | -quant | 1.31B | S | gemma | RedHatAI | 0 | 11.1k |
| tardellirs/aac-board-generator-770m-ptbr | finetune | — | 772M | S | gemma | tardellirs | 0 | 7.5k |
| CraneAILabs/ganda-gemma-1b-GGUF | quantized | GGUF | — | — | gemma | CraneAILabs | 2 | 6.9k |
| DravenBlack/gemma-3-1b-it-Q4_K_M-GGUF | quantized | GGUF | — | — | gemma | DravenBlack | 2 | 6.6k |
| mlx-community/gemma-3-1b-it-qat-8bit | quantized | -8bit | 1.30B | S | gemma | mlx-community | 0 | 6k |
| gaianet/gemma-3-1b-it-GGUF | quantized | GGUF | — | — | gemma | gaianet | 1 | 5.9k |
| mradermacher/gemma-3-1b-it-i1-GGUF | quantized | GGUF | — | — | gemma | mradermacher | 0 | 5.9k |
| Date | Commit | SHA |
|---|---|---|
| 2025-04-04 | Update README.md (#13) | dcc83ea |
| 2025-03-21 | Update README.md (#15) | fa1e42a |
| 2025-03-19 | Adding top_k and top_p to generation_config.json | 417a7e0 |
| 2025-03-12 | Update EOS tokens (#5) | 9b99be8 |
| 2025-03-12 | Update README.md | ed8bfca |
| 2025-03-12 | Update chat_template (#4) | ca8476f |
| 2025-03-12 | Super-squash branch 'main' using huggingface_hub | 6c0ce4e |
Hosted endpoints observed for gemma-3-1b-it, 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 | google/gemma-3-1b-it | — | — | — | — | — | — | — | live | 2026-10-26 |
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: 2026-10-26 · HF API: 2026-10-26 07:15 UTC · Providers: 2026-10-26 07:26 UTC · Articles: 2026-10-26 07:27 UTC · Summary: 2026-10-07 15:00 UTC · Last verified 2026-10-26 07:27 UTC
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
