Choose a creative model with examples from your own work. This list gives candidates and proposed checks, without a prose-quality rank.
We checked the linked vendor sources on October 14, 2026. This post reports no comparative product test.
Check local runtime and model compatibility
These are model candidates for a compatible local runtime. This post does not establish that each model fits an iPhone. Check the exact model file, available memory, and runtime support. These entries do not describe the PocketLLM catalog.
Qwen 2.5, Mistral 7B, Mistral Nemo, Llama 3.2, Phi 3.5 Mini and Gemma 2 for creative writing
1. Qwen2.5-7B-Instruct
Qwen publishes this instruction model under Apache 2.0. See the vendor model card. Ask for two scene outlines with different conflicts. Check that each outline follows your brief.
2. Mistral-7B-Instruct-v0.3
Mistral publishes this instruction model under Apache 2.0. See the vendor model card. Ask for dialogue between named characters. Check each character voice against your notes.
3. Mistral-Nemo-Instruct-2407
Mistral publishes this instruction model under Apache 2.0. See the vendor model card. Provide a chapter outline. Check whether the response preserves events, names, and sequence.
4. Llama-3.2-3B-Instruct
Meta publishes this text model under the Llama 3.2 Community License. See the vendor model card. Ask for a short scene with a fixed viewpoint. Check for changes in viewpoint or tense.
5. Phi-3.5-mini-instruct
Microsoft publishes this instruction model under the MIT license. See the vendor model card. Ask for a scene revision with a specific constraint. Check whether the result follows that constraint.
6. Gemma-2-9b-it
Google describes this model as an English text model under the Gemma terms. See the vendor model card. Use an English passage for an initial check. Compare the revised text with the original voice.
PocketLLM is pre-launch. Private, on-device AI, starting on iPhone, with more platforms planned. No account required, no in-app tracking, and chat processing stays on your device. Join the waitlist.
Qwen, Mistral, Llama, Phi and Gemma: creative text checks
| Option | Proposed text check |
|---|---|
| 1. Qwen2.5-7B-Instruct | Compare plot choices. |
| 2. Mistral-7B-Instruct-v0.3 | Check character voice. |
| 3. Mistral-Nemo-Instruct-2407 | Check continuity. |
| 4. Llama-3.2-3B-Instruct | Check viewpoint. |
| 5. Phi-3.5-mini-instruct | Check the revision constraint. |
| 6. Gemma-2-9b-it | Check English prose. |
Before you choose
- Save the model version and configuration with your sample output.
- Check factual accuracy and compliance with your instructions.
- Compare the output with your original text.
- Select the model only after it passes your own requirements.
Local inference keeps prompts on the device during inference when the selected model runs there. For a local-only workflow, select no cloud model. Disable chat sync and cloud backup. Check the app telemetry policy. Disable optional network features. Review extensions, exports, and local storage access.
Frequently asked questions
Which license does Qwen2.5-7B-Instruct use?
Qwen publishes this instruction model under Apache 2.0. See the Qwen model card.
Which license does Llama-3.2-3B-Instruct use?
Meta publishes this text model under the Llama 3.2 Community License. See the Meta model card.
What language does Google specify for Gemma 2 9B?
Google describes Gemma 2 9B as an English text model under the Gemma terms. See the Google model card.
PocketLLM is pre-launch, with a waitlist for the iPhone launch. More platforms are planned. See the PocketLLM model catalog for its separate model list.