README.md
__init__.py
_tokenizer_identity_snapshot.py
dataset_validator.py
image_geometry.py
pyproject.toml
tokenizer_compat.py
tokenizer_identity.py
./__init__.py
./_tokenizer_identity_snapshot.py
./dataset_validator.py
./image_geometry.py
./tokenizer_compat.py
./tokenizer_identity.py
./case-studies/agentic_rl_text2sql/prepare_data.py
./case-studies/agentic_rl_text2sql/sql_reward.py
./case-studies/agentic_rl_text2sql/sql_tools.py
./case-studies/common/deploy_common.py
./case-studies/common/ep_eval.py
./case-studies/common/eval_common.py
./case-studies/common/figures.py
./case-studies/common/gallery_report.py
./case-studies/common/make_holdout.py
./case-studies/common/study_data.py
./case-studies/common/tests/test_study_data.py
./case-studies/dpo_antislop/antislop_data.py
./case-studies/dpo_antislop/antislop_figures.py
./case-studies/dpo_antislop/antislop_rl.py
./case-studies/dpo_antislop/antislop_tropes.py
./case-studies/dpo_antislop/audit_reward.py
./case-studies/dpo_antislop/build_antislop_pack.py
./case-studies/dpo_antislop/evaluate_antislop.py
./case-studies/dpo_style/make_helpsteer3_dpo.py
./case-studies/embedding_support_search/retrieval_eval.py
./case-studies/grpo_countdown/notebook_helpers.py
./case-studies/lipogram_data/adversarial_probe.py
./case-studies/lipogram_data/audit_unicode.py
./case-studies/lipogram_data/build_rows.py
./case-studies/lipogram_data/checks.py
./case-studies/lipogram_data/lexicon.py
./case-studies/lipogram_data/lipo_eval.py
./case-studies/lipogram_data/pipeline.py
./case-studies/lipogram_data/promote.py
./case-studies/lipogram_data/run_sft.py
./case-studies/lipogram_data/seeds.py
./case-studies/lipogram_data/stats.py
./case-studies/lipogram_data/tests/test_demo_helpers.py
./case-studies/lipogram_data/walkthrough/demo_helpers.py
./case-studies/multilora_fleet/make_massive_shards.py
./case-studies/reasoning_rl/grpo_math_evaluator/test_math.py
./case-studies/rl_ask_before_acting/aba_env.py
./case-studies/rl_ask_before_acting/aba_figures.py
./case-studies/rl_ask_before_acting/ask_before_acting_rl.py
./case-studies/rl_ask_before_acting/evaluate_served.py
./case-studies/rl_ask_before_acting/tests/test_aba_env.py
./case-studies/rl_lipogram/allergy.py
./case-studies/rl_lipogram/assemble_dataset.py
./case-studies/rl_lipogram/build_allergy_eval.py
./case-studies/rl_lipogram/build_prompts.py
./case-studies/rl_lipogram/check_bar.py
./case-studies/rl_lipogram/evaluate_allergy.py
./case-studies/rl_lipogram/evaluate_masked.py
./case-studies/rl_lipogram/evaluate_retention.py
./case-studies/rl_lipogram/evaluate_served.py
./case-studies/rl_lipogram/generate_prompts.py
./case-studies/rl_lipogram/lipogram_data.py
./case-studies/rl_lipogram/lipogram_rewards.py
./case-studies/rl_lipogram/lipogram_rl.py
./case-studies/rl_lipogram/prepare_allergy.py
./case-studies/rl_lipogram/prepare_allergy_v2.py
./case-studies/rl_lipogram/prepare_selfdistill.py
./case-studies/rl_lipogram/prepare_sft.py
./case-studies/rl_lipogram/reward_state.py
./case-studies/rl_lipogram/tools_sneeze_test.py
./case-studies/rl_lipogram/train_sft.py
./case-studies/rl_lipogram/train_trace.py
./case-studies/rl_lipogram/train_trace_allergy.py
./case-studies/rl_lipogram/train_v4.py
./case-studies/rl_lipogram/train_v5.py
./case-studies/rl_lipogram/tests/test_allergy.py
./case-studies/rl_lipogram/tests/test_lipogram_gate.py
./case-studies/rl_lipogram/tests/test_lipogram_rewards.py
./case-studies/rl_lipogram/tests/test_lipogram_training.py
./case-studies/rl_lipogram/tests/test_review_regressions.py
./case-studies/rl_lipogram/walkthrough/build_notebook.py
./case-studies/rl_lipogram/walkthrough/export_recorded.py
./case-studies/rl_lipogram/walkthrough/lipogram_figures.py
./case-studies/sft_bfcl_toolcalls/bfcl_data.py
./case-studies/sft_bfcl_toolcalls/bfcl_score.py
./case-studies/sft_bfcl_toolcalls/build_bfcl_sft.py
./case-studies/sft_caveman/aggregate_seeds.py
./case-studies/sft_caveman/build_caveman_pack.py
./case-studies/sft_caveman/calibrate_judge.py
./case-studies/sft_caveman/caveman_figures.py
./case-studies/sft_caveman/cost_ledger.py
./case-studies/sft_caveman/evaluate_pack.py
./case-studies/sft_caveman/notebook_display.py
./case-studies/sft_caveman/run_pack_sft.py
./case-studies/sft_caveman/validate_data.py
./case-studies/sft_caveman/tests/test_notebook_display.py
./case-studies/sft_humanizer/build_humanizer_pack.py
./case-studies/sft_humanizer/calibrate_judge.py
./case-studies/sft_humanizer/evaluate_pack.py
./case-studies/sft_humanizer/evaluate_walkthrough.py
./case-studies/sft_humanizer/run_pack_sft.py
./case-studies/sft_humanizer/train_walkthrough.py
./case-studies/sft_humanizer/tests/test_checks.py
./case-studies/sft_humanizer/tests/test_walkthrough.py
./case-studies/sft_humanizer/walkthrough/cards.py
./case-studies/sft_humanizer/walkthrough/export_recorded.py
./case-studies/sft_humanizer/walkthrough/humanizer_figures.py
./case-studies/ui_design/author_briefs.py
./case-studies/ui_design/brief_checks.py
./case-studies/ui_design/build_calibration.py
./case-studies/ui_design/build_pack.py
./case-studies/ui_design/compare_arms.py
./case-studies/ui_design/cost_ledger.py
./case-studies/ui_design/evaluate_adapter.py
./case-studies/ui_design/factcheck.py
./case-studies/ui_design/generate.py
./case-studies/ui_design/judge.py
./case-studies/ui_design/judge_design.py
./case-studies/ui_design/render.py
./case-studies/ui_design/repair.py
./case-studies/ui_design/reward.py
./case-studies/ui_design/run_judging.py
./case-studies/ui_design/train_sft.py
./case-studies/ui_design/ui_rl.py
./case-studies/ui_design/tests/test_briefs.py
./case-studies/ui_design/tests/test_harness.py
./case-studies/ui_design/tests/test_network_isolation.py
./dataset_utils/__init__.py
./dataset_utils/image_geometry.py
./dataset_utils/validation.py
./examples/__init__.py
./examples/archive/__init__.py
./examples/archive/vanilla_sampler.py
./examples/archive/countdown/__init__.py
./examples/archive/countdown/countdown_rewards.py
./examples/archive/countdown/countdown_rl.py
./examples/archive/countdown/prepare_data.py
./examples/archive/countdown/data/countdown_train.jsonl
./examples/archive/eval_protocol_chat/__init__.py
./examples/archive/eval_protocol_chat/convert.py
./examples/archive/eval_protocol_chat/reward.py
./examples/archive/eval_protocol_chat/rollout.py
./examples/archive/eval_protocol_chat/train.jsonl
./examples/archive/eval_protocol_chat/train.py
./examples/archive/eval_protocol_chat/remote_server/__init__.py
./examples/archive/eval_protocol_chat/remote_server/server.py
./examples/archive/single_turn_token_in/__init__.py
./examples/archive/single_turn_token_in/rollout.py
./examples/archive/single_turn_token_in/train.py
./examples/decision/__init__.py
./examples/decision/kalshi_v1/__init__.py
./examples/decision/kalshi_v1/prepare_data.py
./examples/decision/kalshi_v1/train_kalshi_decision.py
./examples/decision/kalshi_v1/media/make_split_charts.py
./examples/distillation/gsm8k_privileged/prepare_data.py
./examples/distillation/gsm8k_privileged/train_gsm8k_privileged.py
./examples/distillation/serverless_k3_privileged/train_serverless_k3_privileged.py
./examples/distillation/serverless_k3_privileged/data/gsm8k_200.jsonl
./examples/distillation/serverless_k3_privileged/data/gsm8k_privileged.jsonl
./examples/dpo/__init__.py
./examples/dpo/prepare_data.py
./examples/dpo/train_dpo.py
./examples/dpo/train_dpo_serverless.py
./examples/embedding/retrieval_pairs.jsonl
./examples/embedding/train_embedding.py
./examples/harbor_rl/__init__.py
./examples/harbor_rl/common.py
./examples/harbor_rl/records.py
./examples/harbor_rl/train_opencode.py
./examples/harbor_rl/train_pi.py
./examples/harbor_rl/harnesses/__init__.py
./examples/harbor_rl/harnesses/mimoagent/__init__.py
./examples/harbor_rl/harnesses/mimoagent/agent.py
./examples/harbor_rl/harnesses/mimoagent/constants.py
./examples/harbor_rl/harnesses/mimoagent/driver.py
./examples/harbor_rl/harnesses/mimoagent/mcp_tools.py
./examples/harbor_rl/harnesses/mimoagent/prepare_tasks.py
./examples/harbor_rl/harnesses/mimoagent/rollout.py
./examples/harbor_rl/harnesses/mini_swe/__init__.py
./examples/harbor_rl/harnesses/mini_swe/agent.py
./examples/harbor_rl/harnesses/mini_swe/constants.py
./examples/harbor_rl/harnesses/mini_swe/prepare_tasks.py
./examples/harbor_rl/harnesses/mini_swe/rollout.py
./examples/harbor_rl/harnesses/opencode/__init__.py
./examples/harbor_rl/harnesses/opencode/agent.py
./examples/harbor_rl/harnesses/opencode/artifacts.py
./examples/harbor_rl/harnesses/opencode/config.py
./examples/harbor_rl/harnesses/opencode/constants.py
./examples/harbor_rl/harnesses/opencode/prepare_tasks.py
./examples/harbor_rl/harnesses/opencode/rollout.py
./examples/harbor_rl/harnesses/pi/__init__.py
./examples/harbor_rl/harnesses/pi/agent.py
./examples/harbor_rl/harnesses/pi/artifacts.py
./examples/harbor_rl/harnesses/pi/constants.py
./examples/harbor_rl/harnesses/pi/prepare_tasks.py
./examples/harbor_rl/harnesses/pi/rollout.py
./examples/harbor_rl/tasks/__init__.py
./examples/harbor_rl/tasks/dabstep/__init__.py
./examples/harbor_rl/tasks/dabstep/dataset.py
./examples/harbor_rl/tasks/dabstep/manifest.py
./examples/harbor_rl/tasks/dabstep/prepare_tasks.py
./examples/harbor_rl/tasks/dabstep/train.py
./examples/harbor_rl/tasks/deep_swe/__init__.py
./examples/harbor_rl/tasks/deep_swe/prepare_tasks.py
./examples/harbor_rl/tasks/mimo/__init__.py
./examples/harbor_rl/tasks/mimo/music.py
./examples/harbor_rl/tasks/mimo/sanitize.py
./examples/harbor_rl/tasks/mimo/tasks.py
./examples/harbor_rl/tasks/mimo/assets/__init__.py
./examples/harbor_rl/tasks/mimo/assets/arvo_match.py
./examples/harbor_rl/tasks/mimo/assets/arvo_server.py
./examples/harbor_rl/tasks/mimo/assets/code_collect.py
./examples/harbor_rl/tasks/mimo/assets/code_verify.py
./examples/harbor_rl/tasks/mimo/assets/webdev/__init__.py
./examples/harbor_rl/tasks/mimo/assets/webdev/eval_rubric.py
./examples/harbor_rl/tasks/mimo/assets/webdev/grade_webdev.py
./examples/harbor_rl/tasks/mimo/assets/webdev/shot.py
./examples/harbor_rl/tasks/mimo/mcp/__init__.py
./examples/harbor_rl/tasks/mimo/mcp/client.py
./examples/harbor_rl/tasks/mimo/music_scorer/__init__.py
./examples/harbor_rl/tasks/mimo/music_scorer/core.py
./examples/harbor_rl/tasks/mimo/music_scorer/feats.py
./examples/harbor_rl/tasks/mimo/music_scorer/pipeline.py
./examples/harbor_rl/tasks/mimo/music_scorer/score.py
./examples/harbor_rl/tasks/mimo/music_scorer/baselines/ref_full4k.json
./examples/harbor_rl/tasks/terminal_bench/__init__.py
./examples/harbor_rl/tasks/terminal_bench/train.py
./examples/harbor_rl/tasks/textworld/__init__.py
./examples/harbor_rl/tasks/textworld/dataset.py
./examples/harbor_rl/tasks/textworld/train.py
./examples/harbor_rl/tito/__init__.py
./examples/harbor_rl/tito/_artifact_io.py
./examples/harbor_rl/tito/e2b_frozen.py
./examples/harbor_rl/tito/e2b_templates.py
./examples/harbor_rl/tito/evaluate.py
./examples/harbor_rl/tito/harness_exec.py
./examples/harbor_rl/tito/prepare_tasks.py
./examples/harbor_rl/tito/rollout.py
./examples/harbor_rl/tito/sidecar.py
./examples/harbor_rl/tito/trial.py
./examples/orpo/__init__.py
./examples/orpo/ifeval/train_ifeval_orpo.py
./examples/rl/__init__.py
./examples/rl/deepmath/dataset.jsonl
./examples/rl/deepmath/math_verify_service.py
./examples/rl/deepmath/prepare_data.py
./examples/rl/deepmath/train_deepmath.py
./examples/rl/deepmath/train_deepmath_serverless.py
./examples/rl/experiment/__init__.py
./examples/rl/experiment/sao/__init__.py
./examples/rl/experiment/sao/qwen3_4b_sao.py
./examples/rl/frozen_lake/__init__.py
./examples/rl/frozen_lake/frozen_lake_env.py
./examples/rl/frozen_lake/frozen_lake_rollout.py
./examples/rl/frozen_lake/frozen_lake_schema.py
./examples/rl/frozen_lake/rendering.py
./examples/rl/frozen_lake/seeds.jsonl
./examples/rl/frozen_lake/test_masking.py
./examples/rl/frozen_lake/train_frozen_lake.py
./examples/rl/frozen_lake/verify_rollout.py
./examples/rl/frozen_lake/assets/img/cracked_hole.png
./examples/rl/frozen_lake/assets/img/elf_down.png
./examples/rl/frozen_lake/assets/img/elf_left.png
./examples/rl/frozen_lake/assets/img/elf_right.png
./examples/rl/frozen_lake/assets/img/elf_up.png
./examples/rl/frozen_lake/assets/img/goal.png
./examples/rl/frozen_lake/assets/img/hole.png
./examples/rl/frozen_lake/assets/img/ice.png
./examples/rl/frozen_lake/assets/img/stool.png
./examples/rl/multihop_qa/__init__.py
./examples/rl/multihop_qa/multihop_qa_rollout.py
./examples/rl/multihop_qa/prepare_data.py
./examples/rl/multihop_qa/search_env.py
./examples/rl/multihop_qa/train_multihop_qa_igpo.py
./examples/rl/nemo_gym/__init__.py
./examples/rl/nemo_gym/proxy.py
./examples/rl/nemo_gym/train.py
./examples/rl/visual_toolbench/__init__.py
./examples/rl/visual_toolbench/image_tools.py
./examples/rl/visual_toolbench/reward.py
./examples/rl/visual_toolbench/rollout.py
./examples/rl/visual_toolbench/task.py
./examples/rl/visual_toolbench/train_visual_toolbench.py
./examples/rl/visual_toolbench/train_visual_toolbench_serverless.py
./examples/sft/food_reasoning.jsonl
./examples/sft/muse_glimmer_starter.py
./examples/sft/text2sql_dataset.jsonl
./examples/sft/train_sft.py
./examples/sft/train_sft_serverless.py
./examples/sft/data/support_triage_train.jsonl
./examples/tools/client_phase_trace_demo.py
./examples/tools/create_trainer_then_deployment.py
./examples/tools/list_checkpoints.py
./examples/tools/merge_lora_and_promote.py
./examples/tools/promote_checkpoint.py
./examples/tools/reconnect_and_adjust_lr.py
./examples/tools/verify_logprobs.py
./recipes/async_rl_loop.py
./recipes/decision_loop.py
./recipes/distillation_loop.py
./recipes/dpo_loop.py
./recipes/embedding_loop.py
./recipes/orpo_loop.py
./recipes/rl_loop.py
./recipes/sft_loop.py
./recipes/experiment/__init__.py
./recipes/experiment/async_rl_loop_serverless.py
./recipes/experiment/ppo_value_head_loop.py
./renderer/LICENSE
./renderer/__init__.py
./renderer/_deepseek_v3_split.py
./renderer/_disaggregate_mixin.py
./renderer/_gemma4_split.py
./renderer/_gpt_oss_split.py
./renderer/_image_processing_utils.py
./renderer/_kimi_k25_split.py
./renderer/_kimi_k26_base.py
./renderer/_model_info.py
./renderer/_nemotron3_split.py
./renderer/_qwen3_split.py
./renderer/_registry.py
./renderer/_supervised.py
./renderer/_think_prefill.py
./renderer/_tokenizer_utils.py
./renderer/base.py
./renderer/deepseek_v3.py
./renderer/deepseek_v4.py
./renderer/deepseek_v41.py
./renderer/errors.py
./renderer/gemma4.py
./renderer/glm5.py
./renderer/gpt_oss.py
./renderer/image_fetcher.py
./renderer/image_processing.py
./renderer/kimi_k2.py
./renderer/kimi_k25.py
./renderer/kimi_k26.py
./renderer/kimi_k27_code.py
./renderer/kimi_k2_5_tool_declaration_ts.py
./renderer/kimi_k3.py
./renderer/llama3.py
./renderer/message_weights.py
./renderer/minimax_m2.py
./renderer/minimax_m3.py
./renderer/mistral.py
./renderer/model_info.py
./renderer/muse_glimmer.py
./renderer/muse_spark.py
./renderer/nemotron3.py
./renderer/nemotron35.py
./renderer/plugins.py
./renderer/qwen2_5.py
./renderer/qwen3.py
./renderer/qwen3_5.py
./renderer/reasoning_fields.py
./renderer/role_colon.py
./renderer/supervised.py
./renderer/thinking_trace.py
./renderer/tokenizer.py
./renderer/tito/__init__.py
./renderer/tito/_errors.py
./renderer/tito/deepseek_v4.py
./renderer/tito/deepseek_v41.py
./renderer/tito/glm52.py
./renderer/tito/glm53.py
./renderer/tito/glm53_flash.py
./renderer/tito/kimi_k31.py
./renderer/tito/muse_glimmer.py
./renderer/tito/muse_spark.py
./renderer/tito/plugins.py
./renderer/tito/qwen35.py
./renderer/tito/qwen36.py
./renderer/tito/qwen38.py
./renderer/tito/registry.py
./renderer/tito/shared.py
./renderer/tito/_vendored/__init__.py
./renderer/tito/_vendored/deepseek_v4/LICENSE
./renderer/tito/_vendored/deepseek_v4/PROVENANCE.md
./renderer/tito/_vendored/deepseek_v4/__init__.py
./renderer/tito/_vendored/deepseek_v4/encoding_dsv4.py
./renderer/tito/_vendored/kimi_k31/LICENSE
./renderer/tito/_vendored/kimi_k31/PROVENANCE.md
./renderer/tito/_vendored/kimi_k31/__init__.py
./renderer/tito/_vendored/kimi_k31/encoding_k3.py
./renderer/verifier/__init__.py
./renderer/verifier/__main__.py
./renderer/verifier/cli.py
./renderer/verifier/serve.py
./renderer/verifier/spinup_deployment.py
./renderer/verifier/triage.py
./renderer/verifier/utils/__init__.py
./renderer/verifier/utils/hf_parity.py
./renderer/verifier/utils/inspect.py
./renderer/verifier/utils/inspect_rules.py
./renderer/verifier/utils/presets.py
./renderer/verifier/utils/probe.py
./renderer/verifier/utils/tokenizer.py
./tests/_dataloader_test_helpers.py
./tests/_encoding_dsv41_oracle.LICENSE
./tests/_encoding_dsv41_oracle.py
./tests/_encoding_dsv4_oracle.py
./tests/_tokenizer_identity_snapshot_generator.py
./tests/_v4_tokenizer_data.py
./tests/async_grpo_helpers.py
./tests/conftest.py
./tests/coverage_summary.py
./tests/cpu_recipe_probe.py
./tests/glm5_serverless_cases.py
./tests/test_dataloader_cursor.py
./tests/test_router_replay.py
./tests/test_smoke_imports.py
./tests/verify_installed_artifact.py
./tests/verify_sdk_minimum.py
./tests/e2e/__init__.py
./tests/e2e/conftest.py
./tests/e2e/run_ppo_value_head_convergence.py
./tests/e2e/test_dpo_e2e.py
./tests/e2e/test_dpo_resume_e2e.py
./tests/e2e/test_grpo_e2e.py
./tests/e2e/test_grpo_resume_e2e.py
./tests/e2e/test_pressure_adaptive.py
./tests/e2e/test_reference_e2e.py
./tests/e2e/test_sft_e2e.py
./tests/e2e/test_sft_resume_e2e.py
./tests/fixtures/tito_qwen35_complete_length.json
./tests/smoke_test/__init__.py
./tests/unit/__init__.py
./tests/unit/conftest.py
./tests/unit/renderer_contract.py
./tests/unit/renderer_expected_divergences.py
./tests/unit/renderer_matrix.py
./tests/unit/renderer_multimodal_contract.py
./tests/unit/renderer_scenarios.py
./tests/unit/test_async_rl_evaluation.py
./tests/unit/test_async_rl_loop.py
./tests/unit/test_async_rl_loop_serverless.py
./tests/unit/test_async_rl_telemetry.py
./tests/unit/test_async_rl_train.py
./tests/unit/test_batched_losses.py
./tests/unit/test_checkpoints.py
./tests/unit/test_cispo_loss.py
./tests/unit/test_client.py
./tests/unit/test_client_phase_trace_demo.py
./tests/unit/test_create_trainer_then_deployment.py
./tests/unit/test_cursor_dataloader.py
./tests/unit/test_data_utils.py
./tests/unit/test_dataset_validator.py
./tests/unit/test_decision_calibration.py
./tests/unit/test_decision_loop.py
./tests/unit/test_decision_rows.py
./tests/unit/test_deepseek_v41_renderer.py
./tests/unit/test_deepseek_v4_code_free_dispatch.py
./tests/unit/test_deepseek_v4_renderer.py
./tests/unit/test_disable_thinking_image_weights.py
./tests/unit/test_disaggregate_overrides.py
./tests/unit/test_distillation.py
./tests/unit/test_dpo_loop.py
./tests/unit/test_dppo_loss.py
./tests/unit/test_e2b_frozen.py
./tests/unit/test_embedding_loop.py
./tests/unit/test_eval_protocol_rollout_adapter.py
./tests/unit/test_fileio.py
./tests/unit/test_frozen_lake_base_model_context.py
./tests/unit/test_frozen_lake_verify_validation.py
./tests/unit/test_frozen_lake_visual_rollout.py
./tests/unit/test_gemma4_renderer.py
./tests/unit/test_glm53_renderer.py
./tests/unit/test_glm5_renderer.py
./tests/unit/test_glm_moe_dsa_renderer.py
./tests/unit/test_group_assembler.py
./tests/unit/test_grpo_loss.py
./tests/unit/test_gspo_loss.py
./tests/unit/test_harbor_artifact_io.py
./tests/unit/test_harbor_artifact_process.py
./tests/unit/test_harbor_dabstep.py
./tests/unit/test_harbor_harness_command.py
./tests/unit/test_harbor_harness_exec.py
./tests/unit/test_harbor_mini_swe_exec.py
./tests/unit/test_harbor_sampling.py
./tests/unit/test_harbor_terminal_bench.py
./tests/unit/test_harbor_textworld.py
./tests/unit/test_harbor_tito.py
./tests/unit/test_harbor_tito_plugins.py
./tests/unit/test_heldout_evaluation.py
./tests/unit/test_igpo.py
./tests/unit/test_image_fetcher.py
./tests/unit/test_image_geometry.py
./tests/unit/test_kalshi_prepare_data.py
./tests/unit/test_kimi_history_renderers.py
./tests/unit/test_kimi_k27_code_renderer.py
./tests/unit/test_kimi_k3_renderer.py
./tests/unit/test_llms_txt.py
./tests/unit/test_logging.py
./tests/unit/test_merge_lora_and_promote.py
./tests/unit/test_mimo_code_verifier.py
./tests/unit/test_mimo_harbor_tasks.py
./tests/unit/test_mimo_mcp.py
./tests/unit/test_mimo_sanitize.py
./tests/unit/test_mimoagent_null_device.py
./tests/unit/test_minimax_m2_renderer.py
./tests/unit/test_minimax_m3_renderer.py
./tests/unit/test_mistral_renderer.py
./tests/unit/test_multihop_qa_igpo.py
./tests/unit/test_multimodal_custom_boundary.py
./tests/unit/test_multimodal_target_tokens.py
./tests/unit/test_muse_glimmer_renderer.py
./tests/unit/test_muse_glimmer_starter.py
./tests/unit/test_muse_spark_public_resolution.py
./tests/unit/test_native_cookbook_runtime.py
./tests/unit/test_nemo_gym_failure_diagnostics.py
./tests/unit/test_nemo_gym_proxy.py
./tests/unit/test_nemotron35_public_renderer.py
./tests/unit/test_nemotron3_preserve_thinking.py
./tests/unit/test_orpo_loop.py
./tests/unit/test_orpo_training.py
./tests/unit/test_packaging.py
./tests/unit/test_phase_tracing.py
./tests/unit/test_policy_algorithm.py
./tests/unit/test_policy_anchor.py
./tests/unit/test_ppo_value_head_loop.py
./tests/unit/test_prepare_data.py
./tests/unit/test_progress.py
./tests/unit/test_qwen2_5_renderer.py
./tests/unit/test_qwen3_4b_sao.py
./tests/unit/test_qwen3_6_preserve_thinking.py
./tests/unit/test_qwen3_template_parity.py
./tests/unit/test_qwen_thinking_history.py
./tests/unit/test_recipe_lora_alpha.py
./tests/unit/test_recipe_signal_lifecycle.py
./tests/unit/test_reference.py
./tests/unit/test_renderer_contract_harness.py
./tests/unit/test_renderer_cutover_parity.py
./tests/unit/test_renderer_hf_parity.py
./tests/unit/test_renderer_multimodal_rl.py
./tests/unit/test_renderer_package_registers_everything.py
./tests/unit/test_renderer_plugins.py
./tests/unit/test_renderer_properties.py
./tests/unit/test_resource_autosizing.py
./tests/unit/test_rl_algorithm_validation.py
./tests/unit/test_rl_critic.py
./tests/unit/test_rl_loop.py
./tests/unit/test_rl_metrics_aliases.py
./tests/unit/test_rl_streaming.py
./tests/unit/test_rl_turn_traces.py
./tests/unit/test_rollout_assembler.py
./tests/unit/test_rollout_error_diagnostics.py
./tests/unit/test_rollout_helpers.py
./tests/unit/test_rollout_lifecycle.py
./tests/unit/test_rollout_message.py
./tests/unit/test_rollout_trace.py
./tests/unit/test_rollout_types.py
./tests/unit/test_runner.py
./tests/unit/test_sampling_support.py
./tests/unit/test_sao.py
./tests/unit/test_score_centering_loss.py
./tests/unit/test_sdk_compat.py
./tests/unit/test_sdk_identity_compatibility.py
./tests/unit/test_serverless.py
./tests/unit/test_serverless_countdown.py
./tests/unit/test_serverless_dpo.py
./tests/unit/test_serverless_visual_toolbench.py
./tests/unit/test_service.py
./tests/unit/test_sft_generation_eval.py
./tests/unit/test_sft_loop.py
./tests/unit/test_step_metrics.py
./tests/unit/test_streaming.py
./tests/unit/test_supervised_cutover_parity.py
./tests/unit/test_supervised_rendering.py
./tests/unit/test_sync_rl_batch.py
./tests/unit/test_termination.py
./tests/unit/test_text2sql_train_sft.py
./tests/unit/test_text_datum_construction.py
./tests/unit/test_think_boundary_weights.py
./tests/unit/test_thinking_trace_registry.py
./tests/unit/test_timer.py
./tests/unit/test_tis.py
./tests/unit/test_tito_coverage_metrics.py
./tests/unit/test_tito_deepseek_v4_renderer.py
./tests/unit/test_tito_glm53_flash_renderer.py
./tests/unit/test_tito_glm_null_content.py
./tests/unit/test_tito_harness_adapters.py
./tests/unit/test_tito_kimi_k31_renderer.py
./tests/unit/test_tito_legacy_boundary.py
./tests/unit/test_tito_materializer.py
./tests/unit/test_tito_muse_glimmer_renderer.py
./tests/unit/test_tito_muse_spark_renderer.py
./tests/unit/test_tito_plugins.py
./tests/unit/test_tito_qwen35_renderer.py
./tests/unit/test_tito_qwen36_renderer.py
./tests/unit/test_tito_qwen38_renderer.py
./tests/unit/test_tito_renderer.py
./tests/unit/test_tito_vendored_dsv4_encoder.py
./tests/unit/test_tito_xml_tool_arguments.py
./tests/unit/test_tokenizer_identity_delegation.py
./tests/unit/test_tokenizer_identity_snapshot_generator.py
./tests/unit/test_tokenizers.py
./tests/unit/test_train_deepmath.py
./tests/unit/test_train_frozen_lake.py
./tests/unit/test_training_session.py
./tests/unit/test_turn_matching.py
./tests/unit/test_v4_example_context.py
./tests/unit/test_v4_recipe_context.py
./tests/unit/test_v4_worker_context.py
./tests/unit/test_validate_preflight.py
./tests/unit/test_verifier_presets.py
./tests/unit/test_verifier_probe.py
./tests/unit/test_visual_toolbench.py
./tests/unit/test_weighted_row_invariants.py
./tito/__init__.py
./utils/__init__.py
./utils/checkpoints.py
./utils/client.py
./utils/config.py
./utils/data.py
./utils/dataloader.py
./utils/dataloader_cursor.py
./utils/decision.py
./utils/decision_calibration.py
./utils/embedding.py
./utils/fileio.py
./utils/infra.py
./utils/kimi_k31_tokenizer.py
./utils/logging.py
./utils/losses.py
./utils/memlog.py
./utils/metrics.py
./utils/phase_tracing.py
./utils/progress.py
./utils/resource_autosizing.py
./utils/runner.py
./utils/runner_state.py
./utils/serverless.py
./utils/service.py
./utils/sft_generation_eval.py
./utils/streaming.py
./utils/supervised.py
./utils/termination.py
./utils/timer.py
./utils/tokenizers.py
./utils/training_shapes.py
./utils/validation.py
./utils/distillation/__init__.py
./utils/distillation/eval.py
./utils/distillation/losses.py
./utils/distillation/sampling.py
./utils/distillation/teacher.py
./utils/rl/__init__.py
./utils/rl/anchor.py
./utils/rl/common.py
./utils/rl/critic.py
./utils/rl/heldout.py
./utils/rl/losses.py
./utils/rl/metrics.py
./utils/rl/reference.py
./utils/rl/router_replay.py
./utils/rl/sampling_support.py
./utils/rl/streaming.py
./utils/rl/sync_batch.py
./utils/rl/tis.py
./utils/rl/trial_events.py
./utils/rl/agent/__init__.py
./utils/rl/agent/openai.py
./utils/rl/agent/sampling.py
./utils/rl/agent/session.py
./utils/rl/agent/trajectory.py
./utils/rl/agent/turn_matching.py
./utils/rl/algorithm/__init__.py
./utils/rl/algorithm/cispo.py
./utils/rl/algorithm/dapo.py
./utils/rl/algorithm/dppo.py
./utils/rl/algorithm/dro.py
./utils/rl/algorithm/grpo.py
./utils/rl/algorithm/gspo.py
./utils/rl/algorithm/igpo.py
./utils/rl/algorithm/importance_sampling.py
./utils/rl/algorithm/reinforce.py
./utils/rl/algorithm/sao.py
./utils/rl/algorithm/score_centering.py
./utils/rl/async_rl/__init__.py
./utils/rl/async_rl/batch.py
./utils/rl/async_rl/coordinator.py
./utils/rl/async_rl/errors.py
./utils/rl/async_rl/evaluation.py
./utils/rl/async_rl/producer.py
./utils/rl/async_rl/telemetry.py
./utils/rl/rollout/__init__.py
./utils/rl/rollout/assembler.py
./utils/rl/rollout/eval_protocol.py
./utils/rl/rollout/group_assembler.py
./utils/rl/rollout/lifecycle.py
./utils/rl/rollout/message.py
./utils/rl/rollout/remote.py
./utils/rl/rollout/renderer.py
./utils/rl/rollout/service.py
./utils/rl/rollout/setup.py
./utils/rl/rollout/tito.py
./utils/rl/rollout/trace.py
./utils/rl/rollout/turn_traces.py
./utils/rl/rollout/types.py
case-studies/agentic_rl_text2sql/prepare_data.py
case-studies/agentic_rl_text2sql/sql_reward.py
case-studies/agentic_rl_text2sql/sql_tools.py
case-studies/common/deploy_common.py
case-studies/common/ep_eval.py
case-studies/common/eval_common.py
case-studies/common/figures.py
case-studies/common/gallery_report.py
case-studies/common/make_holdout.py
case-studies/common/study_data.py
case-studies/common/tests/test_study_data.py
case-studies/dpo_antislop/antislop_data.py
case-studies/dpo_antislop/antislop_figures.py
case-studies/dpo_antislop/antislop_rl.py
case-studies/dpo_antislop/antislop_tropes.py
case-studies/dpo_antislop/audit_reward.py
case-studies/dpo_antislop/build_antislop_pack.py
case-studies/dpo_antislop/evaluate_antislop.py
case-studies/dpo_style/make_helpsteer3_dpo.py
case-studies/embedding_support_search/retrieval_eval.py
case-studies/grpo_countdown/notebook_helpers.py
case-studies/lipogram_data/adversarial_probe.py
case-studies/lipogram_data/audit_unicode.py
case-studies/lipogram_data/build_rows.py
case-studies/lipogram_data/checks.py
case-studies/lipogram_data/lexicon.py
case-studies/lipogram_data/lipo_eval.py
case-studies/lipogram_data/pipeline.py
case-studies/lipogram_data/promote.py
case-studies/lipogram_data/run_sft.py
case-studies/lipogram_data/seeds.py
case-studies/lipogram_data/stats.py
case-studies/lipogram_data/tests/test_demo_helpers.py
case-studies/lipogram_data/walkthrough/demo_helpers.py
case-studies/multilora_fleet/make_massive_shards.py
case-studies/reasoning_rl/grpo_math_evaluator/test_math.py
case-studies/rl_ask_before_acting/aba_env.py
case-studies/rl_ask_before_acting/aba_figures.py
case-studies/rl_ask_before_acting/ask_before_acting_rl.py
case-studies/rl_ask_before_acting/evaluate_served.py
case-studies/rl_ask_before_acting/tests/test_aba_env.py
case-studies/rl_lipogram/allergy.py
case-studies/rl_lipogram/assemble_dataset.py
case-studies/rl_lipogram/build_allergy_eval.py
case-studies/rl_lipogram/build_prompts.py
case-studies/rl_lipogram/check_bar.py
case-studies/rl_lipogram/evaluate_allergy.py
case-studies/rl_lipogram/evaluate_masked.py
case-studies/rl_lipogram/evaluate_retention.py
case-studies/rl_lipogram/evaluate_served.py
case-studies/rl_lipogram/generate_prompts.py
case-studies/rl_lipogram/lipogram_data.py
case-studies/rl_lipogram/lipogram_rewards.py
case-studies/rl_lipogram/lipogram_rl.py
case-studies/rl_lipogram/prepare_allergy.py
case-studies/rl_lipogram/prepare_allergy_v2.py
case-studies/rl_lipogram/prepare_selfdistill.py
case-studies/rl_lipogram/prepare_sft.py
case-studies/rl_lipogram/reward_state.py
case-studies/rl_lipogram/tools_sneeze_test.py
case-studies/rl_lipogram/train_sft.py
case-studies/rl_lipogram/train_trace.py
case-studies/rl_lipogram/train_trace_allergy.py
case-studies/rl_lipogram/train_v4.py
case-studies/rl_lipogram/train_v5.py
case-studies/rl_lipogram/tests/test_allergy.py
case-studies/rl_lipogram/tests/test_lipogram_gate.py
case-studies/rl_lipogram/tests/test_lipogram_rewards.py
case-studies/rl_lipogram/tests/test_lipogram_training.py
case-studies/rl_lipogram/tests/test_review_regressions.py
case-studies/rl_lipogram/walkthrough/build_notebook.py
case-studies/rl_lipogram/walkthrough/export_recorded.py
case-studies/rl_lipogram/walkthrough/lipogram_figures.py
case-studies/sft_bfcl_toolcalls/bfcl_data.py
case-studies/sft_bfcl_toolcalls/bfcl_score.py
case-studies/sft_bfcl_toolcalls/build_bfcl_sft.py
case-studies/sft_caveman/aggregate_seeds.py
case-studies/sft_caveman/build_caveman_pack.py
case-studies/sft_caveman/calibrate_judge.py
case-studies/sft_caveman/caveman_figures.py
case-studies/sft_caveman/cost_ledger.py
case-studies/sft_caveman/evaluate_pack.py
case-studies/sft_caveman/notebook_display.py
case-studies/sft_caveman/run_pack_sft.py
case-studies/sft_caveman/validate_data.py
case-studies/sft_caveman/tests/test_notebook_display.py
case-studies/sft_humanizer/build_humanizer_pack.py
case-studies/sft_humanizer/calibrate_judge.py
case-studies/sft_humanizer/evaluate_pack.py
case-studies/sft_humanizer/evaluate_walkthrough.py
case-studies/sft_humanizer/run_pack_sft.py
case-studies/sft_humanizer/train_walkthrough.py
case-studies/sft_humanizer/tests/test_checks.py
case-studies/sft_humanizer/tests/test_walkthrough.py
case-studies/sft_humanizer/walkthrough/cards.py
case-studies/sft_humanizer/walkthrough/export_recorded.py
case-studies/sft_humanizer/walkthrough/humanizer_figures.py
case-studies/ui_design/author_briefs.py
case-studies/ui_design/brief_checks.py
case-studies/ui_design/build_calibration.py
case-studies/ui_design/build_pack.py
case-studies/ui_design/compare_arms.py
case-studies/ui_design/cost_ledger.py
case-studies/ui_design/evaluate_adapter.py
case-studies/ui_design/factcheck.py
case-studies/ui_design/generate.py
case-studies/ui_design/judge.py
case-studies/ui_design/judge_design.py
case-studies/ui_design/render.py
case-studies/ui_design/repair.py
case-studies/ui_design/reward.py
case-studies/ui_design/run_judging.py
case-studies/ui_design/train_sft.py
case-studies/ui_design/ui_rl.py
case-studies/ui_design/tests/test_briefs.py
case-studies/ui_design/tests/test_harness.py
case-studies/ui_design/tests/test_network_isolation.py
dataset_utils/__init__.py
dataset_utils/image_geometry.py
dataset_utils/validation.py
examples/__init__.py
examples/archive/__init__.py
examples/archive/vanilla_sampler.py
examples/archive/countdown/__init__.py
examples/archive/countdown/countdown_rewards.py
examples/archive/countdown/countdown_rl.py
examples/archive/countdown/prepare_data.py
examples/archive/countdown/data/countdown_train.jsonl
examples/archive/eval_protocol_chat/__init__.py
examples/archive/eval_protocol_chat/convert.py
examples/archive/eval_protocol_chat/reward.py
examples/archive/eval_protocol_chat/rollout.py
examples/archive/eval_protocol_chat/train.jsonl
examples/archive/eval_protocol_chat/train.py
examples/archive/eval_protocol_chat/remote_server/__init__.py
examples/archive/eval_protocol_chat/remote_server/server.py
examples/archive/single_turn_token_in/__init__.py
examples/archive/single_turn_token_in/rollout.py
examples/archive/single_turn_token_in/train.py
examples/decision/__init__.py
examples/decision/kalshi_v1/__init__.py
examples/decision/kalshi_v1/prepare_data.py
examples/decision/kalshi_v1/train_kalshi_decision.py
examples/decision/kalshi_v1/media/make_split_charts.py
examples/distillation/gsm8k_privileged/prepare_data.py
examples/distillation/gsm8k_privileged/train_gsm8k_privileged.py
examples/distillation/serverless_k3_privileged/train_serverless_k3_privileged.py
examples/distillation/serverless_k3_privileged/data/gsm8k_200.jsonl
examples/distillation/serverless_k3_privileged/data/gsm8k_privileged.jsonl
examples/dpo/__init__.py
examples/dpo/prepare_data.py
examples/dpo/train_dpo.py
examples/dpo/train_dpo_serverless.py
examples/embedding/retrieval_pairs.jsonl
examples/embedding/train_embedding.py
examples/harbor_rl/__init__.py
examples/harbor_rl/common.py
examples/harbor_rl/records.py
examples/harbor_rl/train_opencode.py
examples/harbor_rl/train_pi.py
examples/harbor_rl/harnesses/__init__.py
examples/harbor_rl/harnesses/mimoagent/__init__.py
examples/harbor_rl/harnesses/mimoagent/agent.py
examples/harbor_rl/harnesses/mimoagent/constants.py
examples/harbor_rl/harnesses/mimoagent/driver.py
examples/harbor_rl/harnesses/mimoagent/mcp_tools.py
examples/harbor_rl/harnesses/mimoagent/prepare_tasks.py
examples/harbor_rl/harnesses/mimoagent/rollout.py
examples/harbor_rl/harnesses/mini_swe/__init__.py
examples/harbor_rl/harnesses/mini_swe/agent.py
examples/harbor_rl/harnesses/mini_swe/constants.py
examples/harbor_rl/harnesses/mini_swe/prepare_tasks.py
examples/harbor_rl/harnesses/mini_swe/rollout.py
examples/harbor_rl/harnesses/opencode/__init__.py
examples/harbor_rl/harnesses/opencode/agent.py
examples/harbor_rl/harnesses/opencode/artifacts.py
examples/harbor_rl/harnesses/opencode/config.py
examples/harbor_rl/harnesses/opencode/constants.py
examples/harbor_rl/harnesses/opencode/prepare_tasks.py
examples/harbor_rl/harnesses/opencode/rollout.py
examples/harbor_rl/harnesses/pi/__init__.py
examples/harbor_rl/harnesses/pi/agent.py
examples/harbor_rl/harnesses/pi/artifacts.py
examples/harbor_rl/harnesses/pi/constants.py
examples/harbor_rl/harnesses/pi/prepare_tasks.py
examples/harbor_rl/harnesses/pi/rollout.py
examples/harbor_rl/tasks/__init__.py
examples/harbor_rl/tasks/dabstep/__init__.py
examples/harbor_rl/tasks/dabstep/dataset.py
examples/harbor_rl/tasks/dabstep/manifest.py
examples/harbor_rl/tasks/dabstep/prepare_tasks.py
examples/harbor_rl/tasks/dabstep/train.py
examples/harbor_rl/tasks/deep_swe/__init__.py
examples/harbor_rl/tasks/deep_swe/prepare_tasks.py
examples/harbor_rl/tasks/mimo/__init__.py
examples/harbor_rl/tasks/mimo/music.py
examples/harbor_rl/tasks/mimo/sanitize.py
examples/harbor_rl/tasks/mimo/tasks.py
examples/harbor_rl/tasks/mimo/assets/__init__.py
examples/harbor_rl/tasks/mimo/assets/arvo_match.py
examples/harbor_rl/tasks/mimo/assets/arvo_server.py
examples/harbor_rl/tasks/mimo/assets/code_collect.py
examples/harbor_rl/tasks/mimo/assets/code_verify.py
examples/harbor_rl/tasks/mimo/assets/webdev/__init__.py
examples/harbor_rl/tasks/mimo/assets/webdev/eval_rubric.py
examples/harbor_rl/tasks/mimo/assets/webdev/grade_webdev.py
examples/harbor_rl/tasks/mimo/assets/webdev/shot.py
examples/harbor_rl/tasks/mimo/mcp/__init__.py
examples/harbor_rl/tasks/mimo/mcp/client.py
examples/harbor_rl/tasks/mimo/music_scorer/__init__.py
examples/harbor_rl/tasks/mimo/music_scorer/core.py
examples/harbor_rl/tasks/mimo/music_scorer/feats.py
examples/harbor_rl/tasks/mimo/music_scorer/pipeline.py
examples/harbor_rl/tasks/mimo/music_scorer/score.py
examples/harbor_rl/tasks/mimo/music_scorer/baselines/ref_full4k.json
examples/harbor_rl/tasks/terminal_bench/__init__.py
examples/harbor_rl/tasks/terminal_bench/train.py
examples/harbor_rl/tasks/textworld/__init__.py
examples/harbor_rl/tasks/textworld/dataset.py
examples/harbor_rl/tasks/textworld/train.py
examples/harbor_rl/tito/__init__.py
examples/harbor_rl/tito/_artifact_io.py
examples/harbor_rl/tito/e2b_frozen.py
examples/harbor_rl/tito/e2b_templates.py
examples/harbor_rl/tito/evaluate.py
examples/harbor_rl/tito/harness_exec.py
examples/harbor_rl/tito/prepare_tasks.py
examples/harbor_rl/tito/rollout.py
examples/harbor_rl/tito/sidecar.py
examples/harbor_rl/tito/trial.py
examples/orpo/__init__.py
examples/orpo/ifeval/train_ifeval_orpo.py
examples/rl/__init__.py
examples/rl/deepmath/dataset.jsonl
examples/rl/deepmath/math_verify_service.py
examples/rl/deepmath/prepare_data.py
examples/rl/deepmath/train_deepmath.py
examples/rl/deepmath/train_deepmath_serverless.py
examples/rl/experiment/__init__.py
examples/rl/experiment/sao/__init__.py
examples/rl/experiment/sao/qwen3_4b_sao.py
examples/rl/frozen_lake/__init__.py
examples/rl/frozen_lake/frozen_lake_env.py
examples/rl/frozen_lake/frozen_lake_rollout.py
examples/rl/frozen_lake/frozen_lake_schema.py
examples/rl/frozen_lake/rendering.py
examples/rl/frozen_lake/seeds.jsonl
examples/rl/frozen_lake/test_masking.py
examples/rl/frozen_lake/train_frozen_lake.py
examples/rl/frozen_lake/verify_rollout.py
examples/rl/frozen_lake/assets/img/cracked_hole.png
examples/rl/frozen_lake/assets/img/elf_down.png
examples/rl/frozen_lake/assets/img/elf_left.png
examples/rl/frozen_lake/assets/img/elf_right.png
examples/rl/frozen_lake/assets/img/elf_up.png
examples/rl/frozen_lake/assets/img/goal.png
examples/rl/frozen_lake/assets/img/hole.png
examples/rl/frozen_lake/assets/img/ice.png
examples/rl/frozen_lake/assets/img/stool.png
examples/rl/multihop_qa/__init__.py
examples/rl/multihop_qa/multihop_qa_rollout.py
examples/rl/multihop_qa/prepare_data.py
examples/rl/multihop_qa/search_env.py
examples/rl/multihop_qa/train_multihop_qa_igpo.py
examples/rl/nemo_gym/__init__.py
examples/rl/nemo_gym/proxy.py
examples/rl/nemo_gym/train.py
examples/rl/visual_toolbench/__init__.py
examples/rl/visual_toolbench/image_tools.py
examples/rl/visual_toolbench/reward.py
examples/rl/visual_toolbench/rollout.py
examples/rl/visual_toolbench/task.py
examples/rl/visual_toolbench/train_visual_toolbench.py
examples/rl/visual_toolbench/train_visual_toolbench_serverless.py
examples/sft/food_reasoning.jsonl
examples/sft/muse_glimmer_starter.py
examples/sft/text2sql_dataset.jsonl
examples/sft/train_sft.py
examples/sft/train_sft_serverless.py
examples/sft/data/support_triage_train.jsonl
examples/tools/client_phase_trace_demo.py
examples/tools/create_trainer_then_deployment.py
examples/tools/list_checkpoints.py
examples/tools/merge_lora_and_promote.py
examples/tools/promote_checkpoint.py
examples/tools/reconnect_and_adjust_lr.py
examples/tools/verify_logprobs.py
fireworks_training_cookbook.egg-info/PKG-INFO
fireworks_training_cookbook.egg-info/SOURCES.txt
fireworks_training_cookbook.egg-info/dependency_links.txt
fireworks_training_cookbook.egg-info/requires.txt
fireworks_training_cookbook.egg-info/top_level.txt
recipes/async_rl_loop.py
recipes/decision_loop.py
recipes/distillation_loop.py
recipes/dpo_loop.py
recipes/embedding_loop.py
recipes/orpo_loop.py
recipes/rl_loop.py
recipes/sft_loop.py
recipes/experiment/__init__.py
recipes/experiment/async_rl_loop_serverless.py
recipes/experiment/ppo_value_head_loop.py
renderer/LICENSE
renderer/__init__.py
renderer/_deepseek_v3_split.py
renderer/_disaggregate_mixin.py
renderer/_gemma4_split.py
renderer/_gpt_oss_split.py
renderer/_image_processing_utils.py
renderer/_kimi_k25_split.py
renderer/_kimi_k26_base.py
renderer/_model_info.py
renderer/_nemotron3_split.py
renderer/_qwen3_split.py
renderer/_registry.py
renderer/_supervised.py
renderer/_think_prefill.py
renderer/_tokenizer_utils.py
renderer/base.py
renderer/deepseek_v3.py
renderer/deepseek_v4.py
renderer/deepseek_v41.py
renderer/errors.py
renderer/gemma4.py
renderer/glm5.py
renderer/gpt_oss.py
renderer/image_fetcher.py
renderer/image_processing.py
renderer/kimi_k2.py
renderer/kimi_k25.py
renderer/kimi_k26.py
renderer/kimi_k27_code.py
renderer/kimi_k2_5_tool_declaration_ts.py
renderer/kimi_k3.py
renderer/llama3.py
renderer/message_weights.py
renderer/minimax_m2.py
renderer/minimax_m3.py
renderer/mistral.py
renderer/model_info.py
renderer/muse_glimmer.py
renderer/muse_spark.py
renderer/nemotron3.py
renderer/nemotron35.py
renderer/plugins.py
renderer/qwen2_5.py
renderer/qwen3.py
renderer/qwen3_5.py
renderer/reasoning_fields.py
renderer/role_colon.py
renderer/supervised.py
renderer/thinking_trace.py
renderer/tokenizer.py
renderer/tito/__init__.py
renderer/tito/_errors.py
renderer/tito/deepseek_v4.py
renderer/tito/deepseek_v41.py
renderer/tito/glm52.py
renderer/tito/glm53.py
renderer/tito/glm53_flash.py
renderer/tito/kimi_k31.py
renderer/tito/muse_glimmer.py
renderer/tito/muse_spark.py
renderer/tito/plugins.py
renderer/tito/qwen35.py
renderer/tito/qwen36.py
renderer/tito/qwen38.py
renderer/tito/registry.py
renderer/tito/shared.py
renderer/tito/_vendored/__init__.py
renderer/tito/_vendored/deepseek_v4/LICENSE
renderer/tito/_vendored/deepseek_v4/PROVENANCE.md
renderer/tito/_vendored/deepseek_v4/__init__.py
renderer/tito/_vendored/deepseek_v4/encoding_dsv4.py
renderer/tito/_vendored/kimi_k31/LICENSE
renderer/tito/_vendored/kimi_k31/PROVENANCE.md
renderer/tito/_vendored/kimi_k31/__init__.py
renderer/tito/_vendored/kimi_k31/encoding_k3.py
renderer/verifier/__init__.py
renderer/verifier/__main__.py
renderer/verifier/cli.py
renderer/verifier/serve.py
renderer/verifier/spinup_deployment.py
renderer/verifier/triage.py
renderer/verifier/utils/__init__.py
renderer/verifier/utils/hf_parity.py
renderer/verifier/utils/inspect.py
renderer/verifier/utils/inspect_rules.py
renderer/verifier/utils/presets.py
renderer/verifier/utils/probe.py
renderer/verifier/utils/tokenizer.py
tests/_dataloader_test_helpers.py
tests/_encoding_dsv41_oracle.LICENSE
tests/_encoding_dsv41_oracle.py
tests/_encoding_dsv4_oracle.py
tests/_tokenizer_identity_snapshot_generator.py
tests/_v4_tokenizer_data.py
tests/async_grpo_helpers.py
tests/conftest.py
tests/coverage_summary.py
tests/cpu_recipe_probe.py
tests/glm5_serverless_cases.py
tests/test_dataloader_cursor.py
tests/test_router_replay.py
tests/test_smoke_imports.py
tests/verify_installed_artifact.py
tests/verify_sdk_minimum.py
tests/e2e/__init__.py
tests/e2e/conftest.py
tests/e2e/run_ppo_value_head_convergence.py
tests/e2e/test_dpo_e2e.py
tests/e2e/test_dpo_resume_e2e.py
tests/e2e/test_grpo_e2e.py
tests/e2e/test_grpo_resume_e2e.py
tests/e2e/test_pressure_adaptive.py
tests/e2e/test_reference_e2e.py
tests/e2e/test_sft_e2e.py
tests/e2e/test_sft_resume_e2e.py
tests/fixtures/tito_qwen35_complete_length.json
tests/smoke_test/__init__.py
tests/unit/__init__.py
tests/unit/conftest.py
tests/unit/renderer_contract.py
tests/unit/renderer_expected_divergences.py
tests/unit/renderer_matrix.py
tests/unit/renderer_multimodal_contract.py
tests/unit/renderer_scenarios.py
tests/unit/test_async_rl_evaluation.py
tests/unit/test_async_rl_loop.py
tests/unit/test_async_rl_loop_serverless.py
tests/unit/test_async_rl_telemetry.py
tests/unit/test_async_rl_train.py
tests/unit/test_batched_losses.py
tests/unit/test_checkpoints.py
tests/unit/test_cispo_loss.py
tests/unit/test_client.py
tests/unit/test_client_phase_trace_demo.py
tests/unit/test_create_trainer_then_deployment.py
tests/unit/test_cursor_dataloader.py
tests/unit/test_data_utils.py
tests/unit/test_dataset_validator.py
tests/unit/test_decision_calibration.py
tests/unit/test_decision_loop.py
tests/unit/test_decision_rows.py
tests/unit/test_deepseek_v41_renderer.py
tests/unit/test_deepseek_v4_code_free_dispatch.py
tests/unit/test_deepseek_v4_renderer.py
tests/unit/test_disable_thinking_image_weights.py
tests/unit/test_disaggregate_overrides.py
tests/unit/test_distillation.py
tests/unit/test_dpo_loop.py
tests/unit/test_dppo_loss.py
tests/unit/test_e2b_frozen.py
tests/unit/test_embedding_loop.py
tests/unit/test_eval_protocol_rollout_adapter.py
tests/unit/test_fileio.py
tests/unit/test_frozen_lake_base_model_context.py
tests/unit/test_frozen_lake_verify_validation.py
tests/unit/test_frozen_lake_visual_rollout.py
tests/unit/test_gemma4_renderer.py
tests/unit/test_glm53_renderer.py
tests/unit/test_glm5_renderer.py
tests/unit/test_glm_moe_dsa_renderer.py
tests/unit/test_group_assembler.py
tests/unit/test_grpo_loss.py
tests/unit/test_gspo_loss.py
tests/unit/test_harbor_artifact_io.py
tests/unit/test_harbor_artifact_process.py
tests/unit/test_harbor_dabstep.py
tests/unit/test_harbor_harness_command.py
tests/unit/test_harbor_harness_exec.py
tests/unit/test_harbor_mini_swe_exec.py
tests/unit/test_harbor_sampling.py
tests/unit/test_harbor_terminal_bench.py
tests/unit/test_harbor_textworld.py
tests/unit/test_harbor_tito.py
tests/unit/test_harbor_tito_plugins.py
tests/unit/test_heldout_evaluation.py
tests/unit/test_igpo.py
tests/unit/test_image_fetcher.py
tests/unit/test_image_geometry.py
tests/unit/test_kalshi_prepare_data.py
tests/unit/test_kimi_history_renderers.py
tests/unit/test_kimi_k27_code_renderer.py
tests/unit/test_kimi_k3_renderer.py
tests/unit/test_llms_txt.py
tests/unit/test_logging.py
tests/unit/test_merge_lora_and_promote.py
tests/unit/test_mimo_code_verifier.py
tests/unit/test_mimo_harbor_tasks.py
tests/unit/test_mimo_mcp.py
tests/unit/test_mimo_sanitize.py
tests/unit/test_mimoagent_null_device.py
tests/unit/test_minimax_m2_renderer.py
tests/unit/test_minimax_m3_renderer.py
tests/unit/test_mistral_renderer.py
tests/unit/test_multihop_qa_igpo.py
tests/unit/test_multimodal_custom_boundary.py
tests/unit/test_multimodal_target_tokens.py
tests/unit/test_muse_glimmer_renderer.py
tests/unit/test_muse_glimmer_starter.py
tests/unit/test_muse_spark_public_resolution.py
tests/unit/test_native_cookbook_runtime.py
tests/unit/test_nemo_gym_failure_diagnostics.py
tests/unit/test_nemo_gym_proxy.py
tests/unit/test_nemotron35_public_renderer.py
tests/unit/test_nemotron3_preserve_thinking.py
tests/unit/test_orpo_loop.py
tests/unit/test_orpo_training.py
tests/unit/test_packaging.py
tests/unit/test_phase_tracing.py
tests/unit/test_policy_algorithm.py
tests/unit/test_policy_anchor.py
tests/unit/test_ppo_value_head_loop.py
tests/unit/test_prepare_data.py
tests/unit/test_progress.py
tests/unit/test_qwen2_5_renderer.py
tests/unit/test_qwen3_4b_sao.py
tests/unit/test_qwen3_6_preserve_thinking.py
tests/unit/test_qwen3_template_parity.py
tests/unit/test_qwen_thinking_history.py
tests/unit/test_recipe_lora_alpha.py
tests/unit/test_recipe_signal_lifecycle.py
tests/unit/test_reference.py
tests/unit/test_renderer_contract_harness.py
tests/unit/test_renderer_cutover_parity.py
tests/unit/test_renderer_hf_parity.py
tests/unit/test_renderer_multimodal_rl.py
tests/unit/test_renderer_package_registers_everything.py
tests/unit/test_renderer_plugins.py
tests/unit/test_renderer_properties.py
tests/unit/test_resource_autosizing.py
tests/unit/test_rl_algorithm_validation.py
tests/unit/test_rl_critic.py
tests/unit/test_rl_loop.py
tests/unit/test_rl_metrics_aliases.py
tests/unit/test_rl_streaming.py
tests/unit/test_rl_turn_traces.py
tests/unit/test_rollout_assembler.py
tests/unit/test_rollout_error_diagnostics.py
tests/unit/test_rollout_helpers.py
tests/unit/test_rollout_lifecycle.py
tests/unit/test_rollout_message.py
tests/unit/test_rollout_trace.py
tests/unit/test_rollout_types.py
tests/unit/test_runner.py
tests/unit/test_sampling_support.py
tests/unit/test_sao.py
tests/unit/test_score_centering_loss.py
tests/unit/test_sdk_compat.py
tests/unit/test_sdk_identity_compatibility.py
tests/unit/test_serverless.py
tests/unit/test_serverless_countdown.py
tests/unit/test_serverless_dpo.py
tests/unit/test_serverless_visual_toolbench.py
tests/unit/test_service.py
tests/unit/test_sft_generation_eval.py
tests/unit/test_sft_loop.py
tests/unit/test_step_metrics.py
tests/unit/test_streaming.py
tests/unit/test_supervised_cutover_parity.py
tests/unit/test_supervised_rendering.py
tests/unit/test_sync_rl_batch.py
tests/unit/test_termination.py
tests/unit/test_text2sql_train_sft.py
tests/unit/test_text_datum_construction.py
tests/unit/test_think_boundary_weights.py
tests/unit/test_thinking_trace_registry.py
tests/unit/test_timer.py
tests/unit/test_tis.py
tests/unit/test_tito_coverage_metrics.py
tests/unit/test_tito_deepseek_v4_renderer.py
tests/unit/test_tito_glm53_flash_renderer.py
tests/unit/test_tito_glm_null_content.py
tests/unit/test_tito_harness_adapters.py
tests/unit/test_tito_kimi_k31_renderer.py
tests/unit/test_tito_legacy_boundary.py
tests/unit/test_tito_materializer.py
tests/unit/test_tito_muse_glimmer_renderer.py
tests/unit/test_tito_muse_spark_renderer.py
tests/unit/test_tito_plugins.py
tests/unit/test_tito_qwen35_renderer.py
tests/unit/test_tito_qwen36_renderer.py
tests/unit/test_tito_qwen38_renderer.py
tests/unit/test_tito_renderer.py
tests/unit/test_tito_vendored_dsv4_encoder.py
tests/unit/test_tito_xml_tool_arguments.py
tests/unit/test_tokenizer_identity_delegation.py
tests/unit/test_tokenizer_identity_snapshot_generator.py
tests/unit/test_tokenizers.py
tests/unit/test_train_deepmath.py
tests/unit/test_train_frozen_lake.py
tests/unit/test_training_session.py
tests/unit/test_turn_matching.py
tests/unit/test_v4_example_context.py
tests/unit/test_v4_recipe_context.py
tests/unit/test_v4_worker_context.py
tests/unit/test_validate_preflight.py
tests/unit/test_verifier_presets.py
tests/unit/test_verifier_probe.py
tests/unit/test_visual_toolbench.py
tests/unit/test_weighted_row_invariants.py
tito/__init__.py
utils/__init__.py
utils/checkpoints.py
utils/client.py
utils/config.py
utils/data.py
utils/dataloader.py
utils/dataloader_cursor.py
utils/decision.py
utils/decision_calibration.py
utils/embedding.py
utils/fileio.py
utils/infra.py
utils/kimi_k31_tokenizer.py
utils/logging.py
utils/losses.py
utils/memlog.py
utils/metrics.py
utils/phase_tracing.py
utils/progress.py
utils/resource_autosizing.py
utils/runner.py
utils/runner_state.py
utils/serverless.py
utils/service.py
utils/sft_generation_eval.py
utils/streaming.py
utils/supervised.py
utils/termination.py
utils/timer.py
utils/tokenizers.py
utils/training_shapes.py
utils/validation.py
utils/distillation/__init__.py
utils/distillation/eval.py
utils/distillation/losses.py
utils/distillation/sampling.py
utils/distillation/teacher.py
utils/rl/__init__.py
utils/rl/anchor.py
utils/rl/common.py
utils/rl/critic.py
utils/rl/heldout.py
utils/rl/losses.py
utils/rl/metrics.py
utils/rl/reference.py
utils/rl/router_replay.py
utils/rl/sampling_support.py
utils/rl/streaming.py
utils/rl/sync_batch.py
utils/rl/tis.py
utils/rl/trial_events.py
utils/rl/agent/__init__.py
utils/rl/agent/openai.py
utils/rl/agent/sampling.py
utils/rl/agent/session.py
utils/rl/agent/trajectory.py
utils/rl/agent/turn_matching.py
utils/rl/algorithm/__init__.py
utils/rl/algorithm/cispo.py
utils/rl/algorithm/dapo.py
utils/rl/algorithm/dppo.py
utils/rl/algorithm/dro.py
utils/rl/algorithm/grpo.py
utils/rl/algorithm/gspo.py
utils/rl/algorithm/igpo.py
utils/rl/algorithm/importance_sampling.py
utils/rl/algorithm/reinforce.py
utils/rl/algorithm/sao.py
utils/rl/algorithm/score_centering.py
utils/rl/async_rl/__init__.py
utils/rl/async_rl/batch.py
utils/rl/async_rl/coordinator.py
utils/rl/async_rl/errors.py
utils/rl/async_rl/evaluation.py
utils/rl/async_rl/producer.py
utils/rl/async_rl/telemetry.py
utils/rl/rollout/__init__.py
utils/rl/rollout/assembler.py
utils/rl/rollout/eval_protocol.py
utils/rl/rollout/group_assembler.py
utils/rl/rollout/lifecycle.py
utils/rl/rollout/message.py
utils/rl/rollout/remote.py
utils/rl/rollout/renderer.py
utils/rl/rollout/service.py
utils/rl/rollout/setup.py
utils/rl/rollout/tito.py
utils/rl/rollout/trace.py
utils/rl/rollout/turn_traces.py
utils/rl/rollout/types.py