Gemma 3 1B IT

Model & attribution

Developer

Google

Parameters

1B

iOS format

GGUF · llama.cpp

Quantization

Q4_K_M

Developed by Google. PocketLLM redistributes the community Q4_K_M GGUF conversion under the Gemma Terms of Use. The required distribution notice is included below.

Original model · iOS artifact source

Publisher’s terms · Prohibited-use policy

Everyday English quality

58.5/100

300 of 300 cases graded. Scores require the complete suite and review.

Quality threshold not met. Reliability validation remains pending.

Three repetition scores: 58.25, 57.50, 59.75.

• Instructions: 40.00/100
• Grounding: 67.50/100
• Conversation: 50.00/100
• Reasoning: 66.67/100
• Writing: 85.00/100
• Honesty: 65.00/100

Executable Python examples: 20/30 responses pass every checked example; 10 fail, 0 extracted payloads do not parse as Python and 0 remain unchecked. Code embedded in prose may require manual review. These finite examples do not prove general coding correctness.

Basic Python coding appendix: 86.7/100 from 30 cases. This rubric score includes partial credit and is separate from everyday quality.

Evaluation method and comparison

License & notices

These are the license texts bundled with PocketLLM for this model. Read or save the plain-text license.

Gemma Terms of Use

Last modified: April 1, 2026

Source: https://ai.google.dev/gemma/terms

================================================================================

Section 1: DEFINITIONS

1.1 Definitions

(a) "Agreement" or "Gemma Terms of Use" means these terms and conditions
    governing use, reproduction, Distribution or modification of Gemma
    Services and incorporated terms.

(b) "Distribution" or "Distribute" means transmission, publication, or
    sharing of Gemma or Model Derivatives to third parties, including via
    hosted service through API, web access, or electronic means ("Hosted
    Service").

(c) "Gemma" means the set of machine learning language models, trained
    weights and parameters identified in the Appendix, regardless of source.

(d) "Google" means Google LLC.

(e) "Model Derivatives" means (i) modifications to Gemma, (ii) works based
    on Gemma, or (iii) machine learning models created by transfer of
    patterns of weights, parameters, operations, or Output of Gemma to
    cause similar performance, including distillation and synthetic data
    methods. Outputs are not Model Derivatives.

(f) "Output" means information content output from Gemma or Model
    Derivative resulting from operation or use.

1.2

"Including" means "including without limitation."

--------------------------------------------------------------------------------

Section 2: ELIGIBILITY AND USAGE

2.1 Eligibility

You represent having legal capacity to enter this Agreement, including
sufficient age of consent. If accessing Gemma Services for a legal entity,
you bind that entity to this Agreement and possess requisite authority.

2.2 Use

You may use, reproduce, modify, Distribute, perform or display Gemma
Services only per this Agreement's terms and must not violate any
provision.

--------------------------------------------------------------------------------

Section 3: DISTRIBUTION AND RESTRICTIONS

3.1 Distribution and Redistribution

You may reproduce or Distribute Gemma or Model Derivatives if meeting all
conditions:

  1. Include use restrictions from Section 3.2 as enforceable provision in
     governing agreements and provide notice that Gemma or Model
     Derivatives are subject to Section 3.2 restrictions.

  2. Provide all third party recipients a copy of this Agreement.

  3. Cause modified files to carry prominent modification notices.

  4. All Distributions (except Hosted Services) must include a "Notice"
     text file stating: "Gemma is provided under and subject to the Gemma
     Terms of Use found at ai.google.dev/gemma/terms."

You may add intellectual property statements to modifications and provide
additional or different terms for modifications or Model Derivatives,
provided use, reproduction, modification, Distribution, performance, and
display comply with this Agreement. Additional terms must not conflict with
Agreement terms.

3.2 Use Restrictions

You must not use Gemma Services:

  1. for restricted uses in the Gemma Prohibited Use Policy at
     ai.google.dev/gemma/prohibited_use_policy (incorporated by reference);
     or

  2. in violation of applicable laws and regulations.

Google reserves the right to restrict usage it reasonably believes violates
this Agreement.

3.3 Generated Output

Google claims no rights in Outputs generated using Gemma. You and users are
solely responsible for Outputs and subsequent uses.

--------------------------------------------------------------------------------

Section 4: ADDITIONAL PROVISIONS

4.1 Updates

Google may update Gemma from time to time.

4.2 Trademarks

Nothing grants rights to use Google's trademarks, trade names, logos or
suggest endorsement. Google reserves unexpressed rights.

4.3 DISCLAIMER OF WARRANTY

"UNLESS REQUIRED BY APPLICABLE LAW, THE GEMMA SERVICES, AND OUTPUTS, ARE
PROVIDED ON AN 'AS IS' BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
EITHER EXPRESS OR IMPLIED, INCLUDING ANY WARRANTIES OR CONDITIONS OF TITLE,
NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE."

You determine appropriateness of use, reproduction, modification,
performance, display or Distribution and assume all associated risks.

4.4 LIMITATION OF LIABILITY

"TO THE FULLEST EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT AND UNDER
NO LEGAL THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), PRODUCT LIABILITY,
CONTRACT, OR OTHERWISE, UNLESS REQUIRED BY APPLICABLE LAW, SHALL GOOGLE OR
ITS AFFILIATES BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT,
INDIRECT, SPECIAL, INCIDENTAL, EXEMPLARY, CONSEQUENTIAL, OR PUNITIVE
DAMAGES, OR LOST PROFITS OF ANY KIND ARISING FROM THIS AGREEMENT OR RELATED
TO, ANY OF THE GEMMA SERVICES OR OUTPUTS."

4.5 Term, Termination, and Survival

This Agreement commences upon acceptance (including through use,
modification, Distribution, reproduction, performance or display of any
Gemma Services portion) and continues until terminated. Google may
terminate for breach. Upon termination, you must delete and cease use and
Distribution of all Gemma and Model Derivatives copies. Sections 1, 2.1,
3.3, and 4.2-4.9 survive termination.

4.6 Governing Law and Jurisdiction

This Agreement is governed by California law without choice of law
principles. UN Convention on Contracts for International Sale of Goods does
not apply. Santa Clara County, California state and federal courts have
exclusive jurisdiction.

4.7 Severability

If any provision is invalid, illegal or unenforceable, remaining provisions
remain valid.

4.8 Entire Agreement

This Agreement states all agreed terms and supersedes other agreements on
its subject matter.

4.9 No Waiver

Google's failure to exercise rights does not constitute waiver of those
rights.

--------------------------------------------------------------------------------

Appendix

Models covered by these Terms of Use:

  - Gemma 1
  - Gemma 1.1
  - Gemma 2
  - Gemma 3
  - Gemma 3n
  - FunctionGemma
  - EmbeddingGemma
  - PaliGemma
  - PaliGemma 2
  - ShieldGemma
  - ShieldGemma 2
  - CodeGemma
  - CodeGemma 1.1
  - Gemma 2 JPN
  - DataGemma RIG
  - DataGemma RAG
  - RecurrentGemma
  - Gemma Scope
  - Gemma-APS
  - T5Gemma
  - VaultGemma
  - T5Gemma 2
  - TranslateGemma

Note: For Gemma 4 terms, see the Gemma 4 license (Apache 2.0). Previous
versions archived at ai.google.dev/gemma/terms-archive.

Gemma Prohibited Use Policy

Last modified: February 21, 2024

Source: https://ai.google.dev/gemma/prohibited_use_policy

================================================================================

You may not use nor allow others to use Gemma or Model Derivatives to:

1. Generate any content, including the outputs or results generated by
   Gemma or Model Derivatives, that infringes, misappropriates, or
   otherwise violates any individual's or entity's rights (including, but
   not limited to rights in copyrighted content).

2. Perform or facilitate dangerous, illegal, or malicious activities,
   including:

   2.1 Facilitation or promotion of illegal activities or violations of
       law, such as:

       2.1.1 Promoting or generating content related to child sexual abuse
             or exploitation;

       2.1.2 Promoting or facilitating sale of, or providing instructions
             for synthesizing or accessing, illegal substances, goods, or
             services;

       2.1.3 Facilitating or encouraging users to commit any type of
             crimes; or

       2.1.4 Promoting or generating violent extremism or terrorist
             content.

   2.2 Engagement in the illegal or unlicensed practice of any vocation or
       profession including, but not limited to, legal, medical,
       accounting, or financial professional practices.

   2.3 Abuse, harm, interference, or disruption of services (or enable
       others to do the same), such as:

       2.3.1 Promoting or facilitating the generation or distribution of
             spam; or

       2.3.2 Generating content for deceptive or fraudulent activities,
             scams, phishing, or malware.

   2.4 Attempts to override or circumvent safety filters or intentionally
       drive Gemma or Model Derivatives to act in a manner that contravenes
       this Gemma Prohibited Use Policy.

   2.5 Generation of content that may harm or promote the harm of
       individuals or a group, such as:

       2.5.1 Generating content that promotes or encourages hatred;

       2.5.2 Facilitating methods of harassment or bullying to intimidate,
             abuse, or insult others;

       2.5.3 Generating content that facilitates, promotes, or incites
             violence;

       2.5.4 Generating content that facilitates, promotes, or encourages
             self harm;

       2.5.5 Generating personally identifying information for distribution
             or other harms;

       2.5.6 Tracking or monitoring people without their consent;

       2.5.7 Generating content that may have unfair or adverse impacts on
             people, particularly impacts related to sensitive or protected
             characteristics; or

       2.5.8 Generating, gathering, processing, or inferring sensitive
             personal or private information about individuals without
             obtaining all rights, authorizations, and consents required
             by applicable laws.

3. Generate and distribute content intended to misinform, misrepresent or
   mislead, including:

   3.1 Misrepresentation of the provenance of generated content by claiming
       content was created by a human, or represent generated content as
       original works, in order to deceive;

   3.2 Generation of content that impersonates an individual (living or
       dead) without explicit disclosure, in order to deceive;

   3.3 Misleading claims of expertise or capability made particularly in
       sensitive areas (e.g. health, finance, government services, or
       legal);

   3.4 Making automated decisions in domains that affect material or
       individual rights or well-being (e.g., finance, legal, employment,
       healthcare, housing, insurance, and social welfare);

   3.5 Generation of defamatory content, including defamatory statements,
       images, or audio content; or

   3.6 Engaging in the unauthorized or unlicensed practice of any
       profession including, but not limited to, financial, legal,
       medical/health, or related professional practices.

4. Generate sexually explicit content, including content created for the
   purposes of pornography or sexual gratification (e.g. sexual chatbots).
   Note that this does not include content created for scientific,
   educational, documentary, or artistic purposes.

Gemma is provided under and subject to the Gemma Terms of Use found at
ai.google.dev/gemma/terms.

PocketLLM redistributes the Gemma 3 1B instruction-tuned model weights
in 4-bit Q4_K_M GGUF quantization under Section 3.1 of the Gemma Terms.

Original model:   google/gemma-3-1b-it (Google DeepMind)
Quantization:     bartowski, 4-bit Q4_K_M GGUF
Upstream source:  https://huggingface.co/lmstudio-community/gemma-3-1b-it-GGUF

Modifications from upstream: the file was quantized to 4-bit Q4_K_M GGUF
format for mobile inference via llama.cpp. No architecture or weight
changes were made beyond quantization. This notice file is bundled as
required by Section 3.1(4) of the Gemma Terms.

For the complete legal text governing use of this model:
  - GemmaTermsOfUse.txt         (Gemma Terms of Use, April 1, 2026)
  - GemmaProhibitedUsePolicy.txt (Gemma Prohibited Use Policy, Feb 21, 2024)

Both files are bundled in this app and are shown to each user during the
first-time model acceptance flow.