careergps_ inference_ lab_ generate
1 MCP server offers this tool
What it usually takes
add_special_tokens– in 1 of 1bad_phrases– in 1 of 1decoding_mode– in 1 of 1diversity_penalty– in 1 of 1early_stopping– in 1 of 1frequency_penalty– in 1 of 1goal– in 1 of 1length_penalty– in 1 of 1max_new_tokens– in 1 of 1max_time– in 1 of 1min_new_tokens– in 1 of 1min_p– in 1 of 1
No server marks it as only reading: treat it as able to change data.
From each publisher's own tool list. We never call a tool.
Servers
| Server | Inputs | Read-only? |
|---|---|---|
CareerGPS Inference Lab co.huggingface.littlesa1gon/careergps-inference-lab Generate CareerGPS text with comprehensive decoding controls. The endpoint supports greedy, sampling, beam, beam-sampling, and contrastive decoding plus repetition, frequency/presence, n-gram, stop-string, bad-phrase, token-suppression, and token-bias controls. It returns text and run diagnostics. | add_special_tokens,bad_phrases,decoding_mode,diversity_penalty,early_stopping,frequency_penalty,goal,length_penalty,max_new_tokens,max_time,min_new_tokens,min_p,no_repeat_ngram_size,num_beam_groups,num_beams,num_return_sequences,penalty_alpha,presence_penalty,prompt,prompt_mode,remove_invalid_values,renormalize_logits,repetition_penalty,seed,skip_special_tokens,stop_sequences,suppress_token_ids,temperature,token_bias_json,top_k,top_p,truncate_prompt_to_fit,typical_p,use_cache | Not stated |