diff --git a/ui/src/types/instance.ts b/ui/src/types/instance.ts
new file mode 100644
index 0000000..4c8ef3c
--- /dev/null
+++ b/ui/src/types/instance.ts
@@ -0,0 +1,181 @@
+export interface Instance {
+  name: string;
+  running: boolean;
+  options?: CreateInstanceOptions;
+}
+
+export interface CreateInstanceOptions {
+
+  auto_restart?: boolean;
+  max_restarts?: number;
+  restart_delay?: number;
+
+  // Llama server options
+  // Common params
+  verbose_prompt?: boolean;
+  threads?: number;
+  threads_batch?: number;
+  cpu_mask?: string;
+  cpu_range?: string;
+  cpu_strict?: number;
+  priority?: number;
+  poll?: number;
+  cpu_mask_batch?: string;
+  cpu_range_batch?: string;
+  cpu_strict_batch?: number;
+  priority_batch?: number;
+  poll_batch?: number;
+  ctx_size?: number;
+  predict?: number;
+  batch_size?: number;
+  ubatch_size?: number;
+  keep?: number;
+  flash_attn?: boolean;
+  no_perf?: boolean;
+  escape?: boolean;
+  no_escape?: boolean;
+  rope_scaling?: string;
+  rope_scale?: number;
+  rope_freq_base?: number;
+  rope_freq_scale?: number;
+  yarn_orig_ctx?: number;
+  yarn_ext_factor?: number;
+  yarn_attn_factor?: number;
+  yarn_beta_slow?: number;
+  yarn_beta_fast?: number;
+  dump_kv_cache?: boolean;
+  no_kv_offload?: boolean;
+  cache_type_k?: string;
+  cache_type_v?: string;
+  defrag_thold?: number;
+  parallel?: number;
+  mlock?: boolean;
+  no_mmap?: boolean;
+  numa?: string;
+  device?: string;
+  override_tensor?: string[];
+  gpu_layers?: number;
+  split_mode?: string;
+  tensor_split?: string;
+  main_gpu?: number;
+  check_tensors?: boolean;
+  override_kv?: string[];
+  lora?: string[];
+  lora_scaled?: string[];
+  control_vector?: string[];
+  control_vector_scaled?: string[];
+  control_vector_layer_range?: string;
+  model?: string;
+  model_url?: string;
+  hf_repo?: string;
+  hf_repo_draft?: string;
+  hf_file?: string;
+  hf_repo_v?: string;
+  hf_file_v?: string;
+  hf_token?: string;
+  log_disable?: boolean;
+  log_file?: string;
+  log_colors?: boolean;
+  verbose?: boolean;
+  verbosity?: number;
+  log_prefix?: boolean;
+  log_timestamps?: boolean;
+
+  // Sampling params
+  samplers?: string;
+  seed?: number;
+  sampling_seq?: string;
+  ignore_eos?: boolean;
+  temperature?: number;
+  top_k?: number;
+  top_p?: number;
+  min_p?: number;
+  xtc_probability?: number;
+  xtc_threshold?: number;
+  typical?: number;
+  repeat_last_n?: number;
+  repeat_penalty?: number;
+  presence_penalty?: number;
+  frequency_penalty?: number;
+  dry_multiplier?: number;
+  dry_base?: number;
+  dry_allowed_length?: number;
+  dry_penalty_last_n?: number;
+  dry_sequence_breaker?: string[];
+  dynatemp_range?: number;
+  dynatemp_exp?: number;
+  mirostat?: number;
+  mirostat_lr?: number;
+  mirostat_ent?: number;
+  logit_bias?: string[];
+  grammar?: string;
+  grammar_file?: string;
+  json_schema?: string;
+  json_schema_file?: string;
+
+  // Server/Example-specific params
+  no_context_shift?: boolean;
+  special?: boolean;
+  no_warmup?: boolean;
+  spm_infill?: boolean;
+  pooling?: string;
+  cont_batching?: boolean;
+  no_cont_batching?: boolean;
+  mmproj?: string;
+  mmproj_url?: string;
+  no_mmproj?: boolean;
+  no_mmproj_offload?: boolean;
+  alias?: string;
+  host?: string;
+  port?: number;
+  path?: string;
+  no_webui?: boolean;
+  embedding?: boolean;
+  reranking?: boolean;
+  api_key?: string;
+  api_key_file?: string;
+  ssl_key_file?: string;
+  ssl_cert_file?: string;
+  chat_template_kwargs?: string;
+  timeout?: number;
+  threads_http?: number;
+  cache_reuse?: number;
+  metrics?: boolean;
+  slots?: boolean;
+  props?: boolean;
+  no_slots?: boolean;
+  slot_save_path?: string;
+  jinja?: boolean;
+  reasoning_format?: string;
+  reasoning_budget?: number;
+  chat_template?: string;
+  chat_template_file?: string;
+  no_prefill_assistant?: boolean;
+  slot_prompt_similarity?: number;
+  lora_init_without_apply?: boolean;
+
+  // Speculative decoding params
+  draft_max?: number;
+  draft_min?: number;
+  draft_p_min?: number;
+  ctx_size_draft?: number;
+  device_draft?: string;
+  gpu_layers_draft?: number;
+  model_draft?: string;
+  cache_type_k_draft?: string;
+  cache_type_v_draft?: string;
+
+  // Audio/TTS params
+  model_vocoder?: string;
+  tts_use_guide_tokens?: boolean;
+
+  // Default model params
+  embd_bge_small_en_default?: boolean;
+  embd_e5_small_en_default?: boolean;
+  embd_gte_small_default?: boolean;
+  fim_qwen_1_5b_default?: boolean;
+  fim_qwen_3b_default?: boolean;
+  fim_qwen_7b_default?: boolean;
+  fim_qwen_7b_spec?: boolean;
+  fim_qwen_14b_spec?: boolean;
+}