Update README.md

README.md

@@ -2,7 +2,7 @@
 base_model: kd13/Type-o1-mini-instruct
 library_name: gguf
 pipeline_tag: text-generation
-license: apache-2.0
+license: mit
 language:
 - en
 - hi
@@ -10,22 +10,22 @@ tags:
 - gguf
 - llama.cpp
 - quantized
-- text-generation
 ---
 
 # Type-o1-mini-instruct - GGUF
 
-GGUF quantizations of [kd13/Type-o1-mini-instruct](https://huggingface.co/kd13/Type-o1-mini-instruct), a compact
-general-purpose instruct model (~1B parameters) for everyday assistant use.
+GGUF quantizations of [kd13/Type-o1-mini-instruct](https://huggingface.co/kd13/Type-o1-mini-instruct), a compact general-purpose instruct model (~1B parameters) for everyday assistant use.
 
 Converted with [llama.cpp](https://github.com/ggml-org/llama.cpp).
-
 The IQ quant was produced with an importance matrix; the rest are static quants.
 
+> **Read the Usage section before running these files.** This model uses a custom chat
+> template, so llama.cpp requires the `--jinja` flag. Without it you will get
+> `this custom template is not supported`.
+
 ## Provided quants
 
-Sorted by size, which is not the same as sorted by quality. IQ-quants are often
-preferable to non-IQ quants of a similar size.
+Sorted by size, which is not the same as sorted by quality. IQ-quants are often preferable to non-IQ quants of a similar size.
 
 | Link | Type | Size/GB | Notes |
 |:-----|:-----|--------:|:------|
@@ -44,33 +44,44 @@ preferable to non-IQ quants of a similar size.
 
 ## Which one should I pick?
 
-For a model this small the practical range is **Q4_K_M through Q8_0**. A 1B model has
-little redundancy to give up, so the very low-bit quants lose more than they would on a
-7B. `Q2_K` and `Q3_K_S` are included for completeness rather than as recommendations.
+For a model this small the practical range is **Q4_K_M through Q8_0**. A 1B model has little redundancy to give up, so the very low-bit quants lose more than they would on a 7B. `Q2_K` and `Q3_K_S` are included for completeness rather than as recommendations.
+
+## Chat template
+
+This model was fine-tuned on a custom template, not the standard Llama 3 header format.
+Each message is wrapped as:
+
+'''
+<|begin_of_text|>{role}
+{content}<|end_of_text|>
+'''
+
+and generation is prompted with a trailing `<|begin_of_text|>assistant\n`. The full Jinja template is embedded in every GGUF file in this repo, so any runtime with Jinja support applies it automatically.
+
+Because this format is not one of llama.cpp's built-in recognised templates, its C++ template matcher will reject it. Passing `--jinja` tells llama.cpp to use the embedded Jinja template instead, which is what you want.
 
 ## Usage
 
-With llama.cpp:
+### llama.cpp
 
 ```bash
-llama-cli -m Type-o1-mini-instruct.Q4_K_M.gguf -cnv
+llama-completion -m Type-o1-mini-instruct.Q4_K_M.gguf --jinja \
+  -sys "You are a helpful assistant." \
+  -p "Explain photosynthesis in two sentences."
 ```
 
-With Ollama:
+Recent llama.cpp builds renamed `llama-cli` to `llama-completion`; on older builds use `llama-cli` with the same flags. For raw text completion with no template applied at all, add `-no-cnv` and drop `--jinja`.
+
+Server:
 
 ```bash
-ollama run hf.co/kd13/Type-o1-mini-instruct-GGUF:Q4_K_M
+llama-server -m Type-o1-mini-instruct.Q4_K_M.gguf --jinja -c 4096
 ```
 
-Also works with LM Studio, koboldcpp, text-generation-webui, and anything else built on
-llama.cpp. If you are new to GGUF files, the llama.cpp README covers the basics
-including how to handle multi-part files.
+Omitting `--jinja` produces `this custom template is not supported` — that is a template-matching error, not a corrupt file.
 
-## Notes and limitations
+### Ollama
 
-Everything in the [base model card](https://huggingface.co/kd13/Type-o1-mini-instruct) applies here.
-In short: this is an experimental small assistant, best suited to general chat, study
-help, writing assistance, summarization, English/Hindi translation, and beginner Python.
-It is not suited to production code generation, non-Python languages, or medical, legal
-or financial decisions. Quantization does not improve any of that - it makes it slightly
-worse in exchange for a smaller file. Verify important outputs and test generated code.
+```bash
+ollama run hf.co/kd13/Type-o1-mini-instruct-GGUF:Q4_K_M
+```