Skip to content

Commit fbcdf6b

Browse files
codelioncodelion
authored
Merge pull request #155 from codelion/codelion-patch-1
Update README.md
2 parents 6663c8f + 4306006 commit fbcdf6b

File tree

1 file changed

+45
-34
lines changed

1 file changed

+45
-34
lines changed

‎README.md

+45-34
Original file line numberDiff line numberDiff line change
@@ -254,40 +254,51 @@ response = client.chat.completions.create(
254254

255255
optillm supports various command-line arguments and environment variables for configuration.
256256

257-
| Parameter | Description | Default Value |
258-
|-------------------------------------|-----------------------------------------------------------------|-----------------|
259-
| `--approach` | Inference approach to use | `"auto"` |
260-
| `--simulations` | Number of MCTS simulations | 2 |
261-
| `--exploration` | Exploration weight for MCTS | 0.2 |
262-
| `--depth` | Simulation depth for MCTS | 1 |
263-
| `--best-of-n` | Number of samples for best_of_n approach | 3 |
264-
| `--model` | OpenAI model to use | `"gpt-4o-mini"` |
265-
| `--base-url` | Base URL for OpenAI compatible endpoint | `""` |
266-
| `--rstar-max-depth` | Maximum depth for rStar algorithm | 3 |
267-
| `--rstar-num-rollouts` | Number of rollouts for rStar algorithm | 5 |
268-
| `--rstar-c` | Exploration constant for rStar algorithm | 1.4 |
269-
| `--n` | Number of final responses to be returned | 1 |
270-
| `--return-full-response` | Return the full response including the CoT with <thinking> tags | `False` |
271-
| `--port` | Specify the port to run the proxy | 8000 |
272-
| `--optillm-api-key` | Optional API key for client authentication to optillm | `""` |
273-
| `--cepo_bestofn_n` | Number of responses to be generated in best of n stage | 3 |
274-
| `--cepo_bestofn_temperature` | Temperature for verifier in best of n stage | 0.1 |
275-
| `--cepo_bestofn_max_tokens` | Maximum number of tokens for verifier in best of n stage | 4096 |
276-
| `--cepo_bestofn_rating_type` | Type of rating in best of n stage ("absolute" or "pairwise") | `"absolute"` |
277-
| `--cepo_planning_n` | Number of plans generated in planning stage | 3 |
278-
| `--cepo_planning_m` | Number of attempts to generate n plans in planning stage | 6 |
279-
| `--cepo_planning_temperature_step1` | Temperature for generator in step 1 of planning stage | 0.55 |
280-
| `--cepo_planning_temperature_step2` | Temperature for generator in step 2 of planning stage | 0.25 |
281-
| `--cepo_planning_temperature_step3` | Temperature for generator in step 3 of planning stage | 0.1 |
282-
| `--cepo_planning_temperature_step4` | Temperature for generator in step 4 of planning stage | 0 |
283-
| `--cepo_planning_max_tokens_step1` | Maximum number of tokens in step 1 of planning stage | 4096 |
284-
| `--cepo_planning_max_tokens_step2` | Maximum number of tokens in step 2 of planning stage | 4096 |
285-
| `--cepo_planning_max_tokens_step3` | Maximum number of tokens in step 3 of planning stage | 4096 |
286-
| `--cepo_planning_max_tokens_step4` | Maximum number of tokens in step 4 of planning stage | 4096 |
287-
| `--cepo_print_output` | Whether to print the output of each stage | `False` |
288-
| `--cepo_config_file` | Path to CePO configuration file | None |
289-
290-
When using Docker, these can be set as environment variables prefixed with `OPTILLM_`.
257+
## Available parameters
258+
259+
optillm supports various command-line arguments for configuration. When using Docker, these can also be set as environment variables prefixed with `OPTILLM_`.
260+
261+
| Parameter | Description | Default Value |
262+
|--------------------------|-----------------------------------------------------------------|-----------------|
263+
| `--approach` | Inference approach to use | `"auto"` |
264+
| `--simulations` | Number of MCTS simulations | 2 |
265+
| `--exploration` | Exploration weight for MCTS | 0.2 |
266+
| `--depth` | Simulation depth for MCTS | 1 |
267+
| `--best-of-n` | Number of samples for best_of_n approach | 3 |
268+
| `--model` | OpenAI model to use | `"gpt-4o-mini"` |
269+
| `--base-url` | Base URL for OpenAI compatible endpoint | `""` |
270+
| `--rstar-max-depth` | Maximum depth for rStar algorithm | 3 |
271+
| `--rstar-num-rollouts` | Number of rollouts for rStar algorithm | 5 |
272+
| `--rstar-c` | Exploration constant for rStar algorithm | 1.4 |
273+
| `--n` | Number of final responses to be returned | 1 |
274+
| `--return-full-response` | Return the full response including the CoT with <thinking> tags | `False` |
275+
| `--port` | Specify the port to run the proxy | 8000 |
276+
| `--optillm-api-key` | Optional API key for client authentication to optillm | `""` |
277+
| `--cepo_*` | See CePO Parameters section below for detailed configuration options | Various |
278+
279+
<details>
280+
<summary><strong>CePO Parameters</strong></summary>
281+
282+
| Parameter | Description | Default Value |
283+
|-----------|-------------|---------------|
284+
| `--cepo_bestofn_n` | Number of responses to be generated in best of n stage | 3 |
285+
| `--cepo_bestofn_temperature` | Temperature for verifier in best of n stage | 0.1 |
286+
| `--cepo_bestofn_max_tokens` | Maximum number of tokens for verifier in best of n stage | 4096 |
287+
| `--cepo_bestofn_rating_type` | Type of rating in best of n stage ("absolute" or "pairwise") | `"absolute"` |
288+
| `--cepo_planning_n` | Number of plans generated in planning stage | 3 |
289+
| `--cepo_planning_m` | Number of attempts to generate n plans in planning stage | 6 |
290+
| `--cepo_planning_temperature_step1` | Temperature for generator in step 1 of planning stage | 0.55 |
291+
| `--cepo_planning_temperature_step2` | Temperature for generator in step 2 of planning stage | 0.25 |
292+
| `--cepo_planning_temperature_step3` | Temperature for generator in step 3 of planning stage | 0.1 |
293+
| `--cepo_planning_temperature_step4` | Temperature for generator in step 4 of planning stage | 0 |
294+
| `--cepo_planning_max_tokens_step1` | Maximum number of tokens in step 1 of planning stage | 4096 |
295+
| `--cepo_planning_max_tokens_step2` | Maximum number of tokens in step 2 of planning stage | 4096 |
296+
| `--cepo_planning_max_tokens_step3` | Maximum number of tokens in step 3 of planning stage | 4096 |
297+
| `--cepo_planning_max_tokens_step4` | Maximum number of tokens in step 4 of planning stage | 4096 |
298+
| `--cepo_print_output` | Whether to print the output of each stage | `False` |
299+
| `--cepo_config_file` | Path to CePO configuration file | None |
300+
301+
</details>
291302

292303
## Running with Docker
293304

0 commit comments

Comments
 (0)