---
title: "Custom forward-backward pass"
method: POST
path: "/rl/training-sessions/{session_id}/operations/custom-forward-backward"
tags: ["RL"]
---

# Custom forward-backward pass

`POST /rl/training-sessions/{session_id}/operations/custom-forward-backward`

Submits a forward-backward pass driven by externally computed gradients of the loss with respect to per-token log-probabilities.

## Path parameters

- `session_id` string, required — Training session ID

## Request body

- RLCustomForwardBackwardBody — Request body for a custom forward-backward pass.
  - `samples` RLTrainingSample[], required — Batch of training samples
    - `model_input` RLModelInput, required
      - `chunks` RLModelInputChunk[], required — Input chunks for the model
        - `encoded_text` RLEncodedTextChunk, required — Pre-tokenized text content for a model input chunk.
          - `tokens` union[], required — Pre-tokenized text input
            - union
              - …
    - `loss_inputs` RLLossInputs, required — Token-level inputs used to compute the loss for one training sample.
      - `target_tokens` RLLossTargetTokens, required
        - `data` union[], required — Integer array of target tokens
          - union
            - string
            - integer
        - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `weights` RLWeights
        - `data` number[], required — Per-token loss weights, one non-negative weight per target token. A weight of 0 excludes the token from loss; fractional weights are honored only by cross-entropy.
        - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `grpo_inputs` RLGRPOLossInputs
        - `advantages` RLLossAdvantages, required
          - `data` number[], required — Float array of per-token advantages
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `logprobs` RLLossLogprobs, required
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `reference_logprobs` RLLossLogprobs
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `importance_sampling_inputs` RLImportanceSamplingLossInputs — Loss inputs for unclipped importance-sampling policy-gradient updates.
        - `advantages` RLLossAdvantages, required
          - `data` number[], required — Float array of per-token advantages
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `logprobs` RLLossLogprobs, required
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `ppo_inputs` RLPPOLossInputs
        - `advantages` RLLossAdvantages, required
          - `data` number[], required — Float array of per-token advantages
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `logprobs` RLLossLogprobs, required
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `cispo_inputs` RLCISPOLossInputs
        - `advantages` RLLossAdvantages, required
          - `data` number[], required — Float array of per-token advantages
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `logprobs` RLLossLogprobs, required
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `dro_inputs` RLDROLossInputs
        - `advantages` RLLossAdvantages, required
          - `data` number[], required — Float array of per-token advantages
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
        - `logprobs` RLLossLogprobs, required
          - `data` number[], required — Float array of per-token log probabilities
          - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'
      - `routed_experts` RLRoutedExperts — Mixture-of-experts routing decisions captured while generating, so training can reuse the same expert selection. A contiguous uint16 buffer of selected expert indices, reshaped by `shape`, which is always `[num_tokens, num_layers, topk]`.
        - `data` string, byte, required — Base64-encoded contiguous uint16 buffer of selected expert indices, row-major over (token, layer, k).
        - `shape` union[], required — Buffer shape as `[num_tokens, num_layers, topk]`.
          - union
            - string
            - integer
  - `gradients` RLTargetLogprobGradients[], required — Per-sample per-token gradients of the loss with respect to log-probabilities
    - `data` number[], required — Float array of per-token gradients (d loss / d log p)
    - `dtype` 'D_TYPE_UNSPECIFIED' | 'D_TYPE_INT64' | 'D_TYPE_FLOAT32' | 'D_TYPE_BFLOAT16'

## Response `200`

- RLCustomForwardBackwardOperation — Async custom forward-backward pass operation
  - `id` string, required — Operation ID
  - `status` 'TRAINING_OPERATION_STATUS_UNSPECIFIED' | 'TRAINING_OPERATION_STATUS_PENDING' | 'TRAINING_OPERATION_STATUS_RUNNING' | 'TRAINING_OPERATION_STATUS_COMPLETED' | 'TRAINING_OPERATION_STATUS_FAILED', required
  - `output` RLCustomForwardBackwardResult — Result of a custom forward-backward pass operation
  - `error` RLTrainingOperationError — Error details for a failed training operation
    - `code` 'TRAINING_OPERATION_ERROR_CODE_UNSPECIFIED' | 'TRAINING_OPERATION_ERROR_CODE_RESOURCE_EXHAUSTED' | 'TRAINING_OPERATION_ERROR_CODE_TIMEOUT' | 'TRAINING_OPERATION_ERROR_CODE_INTERNAL_ERROR' | 'TRAINING_OPERATION_ERROR_CODE_SESSION_NOT_ACTIVE' | 'TRAINING_OPERATION_ERROR_CODE_INVALID_INPUT' | 'TRAINING_OPERATION_ERROR_CODE_NON_FINITE_LOSS' — Application error code for a failed training operation
    - `message` string — Human-readable error message

## Other responses

- `default` — An unexpected error response.

---

[API](https://skmtc.net/together/apis/together-apis.md) · [All operations](https://skmtc.net/together/apis/together-apis/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/together/together-apis/revisions/f26a23fab8e0/schema)