diff --git a/docs/BENCHMARKS.md b/docs/BENCHMARKS.md
index 3e6076e212611fa15eacc659522e7e732b0c6456..918c53ac3606096afc487a65e5315aace55386c3 100644
--- a/docs/BENCHMARKS.md
+++ b/docs/BENCHMARKS.md
@@ -53,8 +53,8 @@ To ensure benchmark integrity, follow these steps when adding or updating result
#### 3. Record Scores & Plots
* **Score**: At run completion, extract `total_reward_ma` from logs (`trial_metrics`).
-* **Link**: Add HuggingFace folder link: `[FOLDER](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/FOLDER)`
-* **Pull data**: `source .env && uv run hf download SLM-Lab/benchmark-dev --include "data/FOLDER/*" --local-dir hf_data --repo-type dataset`
+* **Link**: Add HuggingFace folder link: `[FOLDER](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/FOLDER)`
+* **Pull data**: `source .env && uv run hf download SLM-Lab/benchmark --include "data/FOLDER/*" --local-dir hf_data --repo-type dataset`
* **Plot**: Generate with folders from table:
```bash
slm-lab plot -t "CartPole-v1" -f ppo_cartpole_2026...,dqn_cartpole_2026...
@@ -114,7 +114,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| 1 | Classic Control | 3 | 🔄 | 🔄 | 🔄 | 🔄 | 🔄 | 🔄 | 🔄 | Rerun pending |
| 2 | Box2D | 2 | N/A | N/A | 🔄 | 🔄 | 🔄 | 🔄 | 🔄 | Rerun pending |
| 3 | MuJoCo | 11 | N/A | N/A | N/A | N/A | 🔄 | 🔄 | 🔄 | Rerun pending |
-| 4 | Atari | 59 | N/A | N/A | Skip | Skip | Skip | 🔄 | N/A | **54 games** (not in this rerun) |
+| 4 | Atari | 59 | N/A | N/A | N/A | Skip | 🔄 | ✅ | N/A | **54 games** |
**Legend**: ✅ Solved | ⚠️ Close (>80%) | 📊 Acceptable | ❌ Failed | 🔄 In progress/Pending | Skip Not started | N/A Not applicable
@@ -140,7 +140,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| PPO | ✅ | 499.54 | [slm_lab/spec/benchmark/ppo/ppo_cartpole.json](../slm_lab/spec/benchmark/ppo/ppo_cartpole.json) | ppo_cartpole | [ppo_cartpole_2026_01_30_221924](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_cartpole_2026_01_30_221924) |
| SAC | ⚠️ | 359.69 | [slm_lab/spec/benchmark/sac/sac_cartpole.json](../slm_lab/spec/benchmark/sac/sac_cartpole.json) | sac_cartpole | [sac_cartpole_2026_01_30_221934](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/sac_cartpole_2026_01_30_221934) |
-
+
#### 1.2 Acrobot-v1
@@ -156,7 +156,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| PPO | ✅ | -81.43 | [slm_lab/spec/benchmark/ppo/ppo_acrobot.json](../slm_lab/spec/benchmark/ppo/ppo_acrobot.json) | ppo_acrobot | [ppo_acrobot_2026_01_30_215352](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_acrobot_2026_01_30_215352) |
| SAC | ✅ | -97.08 | [slm_lab/spec/benchmark/sac/sac_acrobot.json](../slm_lab/spec/benchmark/sac/sac_acrobot.json) | sac_acrobot | [sac_acrobot_2026_01_30_215401](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/sac_acrobot_2026_01_30_215401) |
-
+
#### 1.3 Pendulum-v1
@@ -170,7 +170,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| PPO | ✅ | -168.26 | [slm_lab/spec/benchmark/ppo/ppo_pendulum.json](../slm_lab/spec/benchmark/ppo/ppo_pendulum.json) | ppo_pendulum | [ppo_pendulum_2026_01_30_215944](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_pendulum_2026_01_30_215944) |
| SAC | ✅ | -152.33 | [slm_lab/spec/benchmark/sac/sac_pendulum.json](../slm_lab/spec/benchmark/sac/sac_pendulum.json) | sac_pendulum | [sac_pendulum_2026_01_30_215454](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/sac_pendulum_2026_01_30_215454) |
-
+
### Phase 2: Box2D
@@ -188,7 +188,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| PPO | ⚠️ | 159.02 | [slm_lab/spec/benchmark/ppo/ppo_lunar.json](../slm_lab/spec/benchmark/ppo/ppo_lunar.json) | ppo_lunar | [ppo_lunar_2026_01_30_215550](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_lunar_2026_01_30_215550) |
| SAC | ❌ | -75.43 | [slm_lab/spec/benchmark/sac/sac_lunar.json](../slm_lab/spec/benchmark/sac/sac_lunar.json) | sac_lunar | [sac_lunar_2026_01_30_215552](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/sac_lunar_2026_01_30_215552) |
-
+
#### 2.2 LunarLander-v3 (Continuous)
@@ -202,7 +202,7 @@ Search budget: ~3-4 trials per dimension (8 trials = 2-3 dims, 16 = 3-4 dims, 20
| PPO | ⚠️ | 165.48 | [slm_lab/spec/benchmark/ppo/ppo_lunar.json](../slm_lab/spec/benchmark/ppo/ppo_lunar.json) | ppo_lunar_continuous | [ppo_lunar_continuous_2026_01_31_104549](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_lunar_continuous_2026_01_31_104549) |
| SAC | ✅ | 208.60 | [slm_lab/spec/benchmark/sac/sac_lunar.json](../slm_lab/spec/benchmark/sac/sac_lunar.json) | sac_lunar_continuous | [sac_lunar_continuous_2026_01_31_104537](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/sac_lunar_continuous_2026_01_31_104537) |
-
+
### Phase 3: MuJoCo
@@ -247,15 +247,15 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
#### 3.1 Hopper-v5
-**Docs**: [Hopper](https://gymnasium.farama.org/environments/mujoco/hopper/) | State: Box(11) | Action: Box(3) | Target reward MA > 2000
+**Docs**: [Hopper](https://gymnasium.farama.org/environments/mujoco/hopper/) | State: Box(11) | Action: Box(3) | Target reward MA ~ 2000
**Settings**: max_frame 4e6 | num_envs 16 | max_session 4 | log_frequency 1e4
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ⚠️ | 1174.57 | [slm_lab/spec/benchmark/ppo/ppo_hopper.json](../slm_lab/spec/benchmark/ppo/ppo_hopper.json) | ppo_hopper | [ppo_hopper_2026_01_30_220138](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_hopper_2026_01_30_220138) |
+| PPO | ✅ | 1972.38 | [slm_lab/spec/benchmark/ppo/ppo_hopper.json](../slm_lab/spec/benchmark/ppo/ppo_hopper.json) | ppo_hopper | [ppo_hopper_2026_01_31_105438](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_hopper_2026_01_31_105438) |
-
+
#### 3.2 HalfCheetah-v5
@@ -265,9 +265,9 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 5851.70 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_halfcheetah_2026_01_30_230302](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_halfcheetah_2026_01_30_230302) |
+| PPO | ✅ | 5851.70 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_halfcheetah_2026_01_30_230302](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_halfcheetah_2026_01_30_230302) |
-
+
#### 3.3 Walker2d-v5
@@ -277,9 +277,9 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 4042.07 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_walker2d_2026_01_30_222124](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_walker2d_2026_01_30_222124) |
+| PPO | ✅ | 4042.07 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_walker2d_2026_01_30_222124](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_walker2d_2026_01_30_222124) |
-
+
#### 3.4 Ant-v5
@@ -289,9 +289,9 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 2514.64 | [slm_lab/spec/benchmark/ppo/ppo_ant.json](../slm_lab/spec/benchmark/ppo/ppo_ant.json) | ppo_ant | [ppo_ant_2026_01_31_042006](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_ant_2026_01_31_042006) |
+| PPO | ✅ | 2514.64 | [slm_lab/spec/benchmark/ppo/ppo_ant.json](../slm_lab/spec/benchmark/ppo/ppo_ant.json) | ppo_ant | [ppo_ant_2026_01_31_042006](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_ant_2026_01_31_042006) |
-
+
#### 3.5 Swimmer-v5
@@ -301,21 +301,21 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 229.31 | [slm_lab/spec/benchmark/ppo/ppo_swimmer.json](../slm_lab/spec/benchmark/ppo/ppo_swimmer.json) | ppo_swimmer | [ppo_swimmer_2026_01_30_215922](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_swimmer_2026_01_30_215922) |
+| PPO | ✅ | 229.31 | [slm_lab/spec/benchmark/ppo/ppo_swimmer.json](../slm_lab/spec/benchmark/ppo/ppo_swimmer.json) | ppo_swimmer | [ppo_swimmer_2026_01_30_215922](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_swimmer_2026_01_30_215922) |
-
+
#### 3.6 Reacher-v5
-**Docs**: [Reacher](https://gymnasium.farama.org/environments/mujoco/reacher/) | State: Box(11) | Action: Box(2) | Target reward MA > -10
+**Docs**: [Reacher](https://gymnasium.farama.org/environments/mujoco/reacher/) | State: Box(10) | Action: Box(2) | Target reward MA > -10
**Settings**: max_frame 4e6 | num_envs 16 | max_session 4 | log_frequency 1e4
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | -5.08 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco_longhorizon | [ppo_mujoco_longhorizon_reacher_2026_01_30_215805](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_longhorizon_reacher_2026_01_30_215805) |
+| PPO | ✅ | -5.08 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco_longhorizon | [ppo_mujoco_longhorizon_reacher_2026_01_30_215805](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_longhorizon_reacher_2026_01_30_215805) |
-
+
#### 3.7 Pusher-v5
@@ -325,9 +325,9 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | -49.09 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco_longhorizon | [ppo_mujoco_longhorizon_pusher_2026_01_30_215824](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_longhorizon_pusher_2026_01_30_215824) |
+| PPO | ✅ | -49.09 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco_longhorizon | [ppo_mujoco_longhorizon_pusher_2026_01_30_215824](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_longhorizon_pusher_2026_01_30_215824) |
-
+
#### 3.8 InvertedPendulum-v5
@@ -337,77 +337,206 @@ source .env && slm-lab run-remote --gpu -s max_frame=MAX_FRAME \
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 944.87 | [slm_lab/spec/benchmark/ppo/ppo_inverted_pendulum.json](../slm_lab/spec/benchmark/ppo/ppo_inverted_pendulum.json) | ppo_inverted_pendulum | [ppo_inverted_pendulum_2026_01_30_230211](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_inverted_pendulum_2026_01_30_230211) |
+| PPO | ✅ | 944.87 | [slm_lab/spec/benchmark/ppo/ppo_inverted_pendulum.json](../slm_lab/spec/benchmark/ppo/ppo_inverted_pendulum.json) | ppo_inverted_pendulum | [ppo_inverted_pendulum_2026_01_30_230211](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_inverted_pendulum_2026_01_30_230211) |
-
+
#### 3.9 InvertedDoublePendulum-v5
-**Docs**: [InvertedDoublePendulum](https://gymnasium.farama.org/environments/mujoco/inverted_double_pendulum/) | State: Box(11) | Action: Box(1) | Target reward MA ~8000
+**Docs**: [InvertedDoublePendulum](https://gymnasium.farama.org/environments/mujoco/inverted_double_pendulum/) | State: Box(9) | Action: Box(1) | Target reward MA ~8000
**Settings**: max_frame 10e6 | num_envs 16 | max_session 4 | log_frequency 1e4
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 7622.00 | [slm_lab/spec/benchmark/ppo/ppo_inverted_double_pendulum.json](../slm_lab/spec/benchmark/ppo/ppo_inverted_double_pendulum.json) | ppo_inverted_double_pendulum | [ppo_inverted_double_pendulum_2026_01_30_220651](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_inverted_double_pendulum_2026_01_30_220651) |
+| PPO | ✅ | 7622.00 | [slm_lab/spec/benchmark/ppo/ppo_inverted_double_pendulum.json](../slm_lab/spec/benchmark/ppo/ppo_inverted_double_pendulum.json) | ppo_inverted_double_pendulum | [ppo_inverted_double_pendulum_2026_01_30_220651](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_inverted_double_pendulum_2026_01_30_220651) |
-
+
#### 3.10 Humanoid-v5
-**Docs**: [Humanoid](https://gymnasium.farama.org/environments/mujoco/humanoid/) | State: Box(376) | Action: Box(17) | Target reward MA > 1000
+**Docs**: [Humanoid](https://gymnasium.farama.org/environments/mujoco/humanoid/) | State: Box(348) | Action: Box(17) | Target reward MA > 1000
**Settings**: max_frame 10e6 | num_envs 16 | max_session 4 | log_frequency 1e4
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 3774.08 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_humanoid_2026_01_30_222339](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_humanoid_2026_01_30_222339) |
+| PPO | ✅ | 3774.08 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_humanoid_2026_01_30_222339](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_humanoid_2026_01_30_222339) |
-
+
#### 3.11 HumanoidStandup-v5
-**Docs**: [HumanoidStandup](https://gymnasium.farama.org/environments/mujoco/humanoid_standup/) | State: Box(376) | Action: Box(17) | Target reward MA > 100000
+**Docs**: [HumanoidStandup](https://gymnasium.farama.org/environments/mujoco/humanoid_standup/) | State: Box(348) | Action: Box(17) | Target reward MA > 100000
**Settings**: max_frame 4e6 | num_envs 16 | max_session 4 | log_frequency 1e4
| Algorithm | Status | MA | SPEC_FILE | SPEC_NAME | HF Repo |
|-----------|--------|-----|-----------|-----------|---------|
-| PPO | ✅ | 165841.17 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_humanoidstandup_2026_01_30_215802](https://huggingface.co/datasets/SLM-Lab/benchmark-dev/tree/main/data/ppo_mujoco_humanoidstandup_2026_01_30_215802) |
+| PPO | ✅ | 165841.17 | [slm_lab/spec/benchmark/ppo/ppo_mujoco.json](../slm_lab/spec/benchmark/ppo/ppo_mujoco.json) | ppo_mujoco | [ppo_mujoco_humanoidstandup_2026_01_30_215802](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_mujoco_humanoidstandup_2026_01_30_215802) |
-
+
### Phase 4: Atari
-**Docs**: [Atari environments](https://ale.farama.org/environments/) | State: Box(84,84,4 after preprocessing) | Action: Discrete(4-18, game-dependent) | Solved: Game-specific thresholds
+**Docs**: [Atari environments](https://ale.farama.org/environments/) | State: Box(84,84,4 after preprocessing) | Action: Discrete(4-18, game-dependent)
**Settings**: max_frame 10e6 | num_envs 16 | max_session 4 | log_frequency 10000
**Environment**:
- Gymnasium ALE v5 with `life_loss_info=true`
-- v5 is harder than v4 due to sticky actions (default `repeat_action_probability=0.25` vs v4's 0.0), which randomly repeats agent actions to simulate console stochasticity and prevent memorization, following [Machado et al. (2018)](https://arxiv.org/abs/1709.06009) research best practices. See [ALE version history](https://ale.farama.org/environments/#version-history-and-naming-schemes).
+- v5 uses sticky actions (`repeat_action_probability=0.25`) per [Machado et al. (2018)](https://arxiv.org/abs/1709.06009) best practices
-**Algorithm: PPO**:
-- **Network**: ConvNet [32,64,64] + 512fc (Nature CNN), orthogonal init, normalize=true, clip_grad_val=0.5
-- **Hyperparams**: AdamW (lr=2.5e-4, eps=1e-5), minibatch_size=256, time_horizon=128, training_epoch=4, clip_eps=0.1, entropy_coef=0.01
+**Algorithm Specs** (all use Nature CNN [32,64,64] + 512fc):
+- **DDQN+PER**: Skipped - off-policy variants ~6x slower (~230 fps vs ~1500 fps), not cost effective at 10M frames
+- **A2C**: [a2c_gae_atari.json](../slm_lab/spec/benchmark/a2c/a2c_gae_atari.json) - RMSprop (lr=7e-4), training_frequency=32
+- **PPO**: [ppo_atari.json](../slm_lab/spec/benchmark/ppo/ppo_atari.json) - AdamW (lr=2.5e-4), minibatch=256, horizon=128, epochs=4
-**Lambda Variants**: All use one spec file ([slm_lab/spec/benchmark/ppo/ppo_atari.json](../slm_lab/spec/benchmark/ppo/ppo_atari.json)), differing only in GAE lambda. Lower lambda = bias toward immediate rewards (action games), higher = longer credit horizon (strategic games).
+**PPO Lambda Variants** (table shows best result per game):
| SPEC_NAME | Lambda | Best for |
|-----------|--------|----------|
-| ppo_atari | 0.95 | Long-horizon, strategic games (default) |
-| ppo_atari_lam85 | 0.85 | Mixed/moderate games |
-| ppo_atari_lam70 | 0.70 | Fast action games |
+| ppo_atari | 0.95 | Strategic games (default) |
+| ppo_atari_lam85 | 0.85 | Mixed games |
+| ppo_atari_lam70 | 0.70 | Action games |
-**Reproduce**: Copy `ENV` from first column, `SPEC_NAME` from column header. All use the same SPEC_FILE:
+**Reproduce**:
```bash
-source .env && slm-lab run-remote --gpu -s env=ENV \
+# A2C
+source .env && slm-lab run-remote --gpu -s env=ENV -s max_frame=1e7 \
+ slm_lab/spec/benchmark/a2c/a2c_gae_atari.json a2c_gae_atari train -n NAME
+
+# PPO
+source .env && slm-lab run-remote --gpu -s env=ENV -s max_frame=1e7 \
slm_lab/spec/benchmark/ppo/ppo_atari.json SPEC_NAME train -n NAME
```
-| ENV\SPEC_NAME | ppo_atari | ppo_atari_lam85 | ppo_atari_lam70 |
-| -------- | ----------------- | --------------- | --------------- |
-| ALE/Adventure-v5 | Skip | Skip | Skip |
+| ENV | Score | SPEC_NAME | HF Repo |
+|-----|-------|-----------|---------|
+| ALE/AirRaid-v5 | 5067 | a2c_gae_atari | [a2c_gae_atari_airraid_2026_02_01_082446](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_airraid_2026_02_01_082446) |
+| ALE/AirRaid-v5 | 8245 | ppo_atari | [ppo_atari_airraid_2026_01_06_113119](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_airraid_2026_01_06_113119) |
+| ALE/Alien-v5 | 1488 | a2c_gae_atari | [a2c_gae_atari_alien_2026_02_01_000858](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_alien_2026_02_01_000858) |
+| ALE/Alien-v5 | 1453 | ppo_atari | [ppo_atari_alien_2026_01_06_112514](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_alien_2026_01_06_112514) |
+| ALE/Amidar-v5 | 330 | a2c_gae_atari | [a2c_gae_atari_amidar_2026_02_01_082251](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_amidar_2026_02_01_082251) |
+| ALE/Amidar-v5 | 580 | ppo_atari_lam85 | [ppo_atari_lam85_amidar_2026_01_07_223416](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_amidar_2026_01_07_223416) |
+| ALE/Assault-v5 | 1646 | a2c_gae_atari | [a2c_gae_atari_assault_2026_02_01_082252](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_assault_2026_02_01_082252) |
+| ALE/Assault-v5 | 4293 | ppo_atari_lam85 | [ppo_atari_lam85_assault_2026_01_08_130044](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_assault_2026_01_08_130044) |
+| ALE/Asterix-v5 | 2712 | a2c_gae_atari | [a2c_gae_atari_asterix_2026_02_01_082315](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_asterix_2026_02_01_082315) |
+| ALE/Asterix-v5 | 3482 | ppo_atari_lam85 | [ppo_atari_lam85_asterix_2026_01_07_223445](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_asterix_2026_01_07_223445) |
+| ALE/Asteroids-v5 | 2106 | a2c_gae_atari | [a2c_gae_atari_asteroids_2026_02_01_082328](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_asteroids_2026_02_01_082328) |
+| ALE/Asteroids-v5 | 1554 | ppo_atari_lam85 | [ppo_atari_lam85_asteroids_2026_01_07_224245](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_asteroids_2026_01_07_224245) |
+| ALE/Atlantis-v5 | 873365 | a2c_gae_atari | [a2c_gae_atari_atlantis_2026_02_01_082330](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_atlantis_2026_02_01_082330) |
+| ALE/Atlantis-v5 | 792886 | ppo_atari | [ppo_atari_atlantis_2026_01_06_120440](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_atlantis_2026_01_06_120440) |
+| ALE/BankHeist-v5 | 1099 | a2c_gae_atari | [a2c_gae_atari_bankheist_2026_02_01_082403](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_bankheist_2026_02_01_082403) |
+| ALE/BankHeist-v5 | 1045 | ppo_atari | [ppo_atari_bankheist_2026_01_06_121042](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_bankheist_2026_01_06_121042) |
+| ALE/BattleZone-v5 | 2437 | a2c_gae_atari | [a2c_gae_atari_battlezone_2026_02_01_082425](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_battlezone_2026_02_01_082425) |
+| ALE/BattleZone-v5 | 26383 | ppo_atari_lam85 | [ppo_atari_lam85_battlezone_2026_01_08_094729](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_battlezone_2026_01_08_094729) |
+| ALE/BeamRider-v5 | 2767 | a2c_gae_atari | [a2c_gae_atari_beamrider_2026_02_01_000921](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_beamrider_2026_02_01_000921) |
+| ALE/BeamRider-v5 | 2765 | ppo_atari | [ppo_atari_beamrider_2026_01_06_112533](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_beamrider_2026_01_06_112533) |
+| ALE/Berzerk-v5 | 439 | a2c_gae_atari | [a2c_gae_atari_berzerk_2026_02_01_082540](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_berzerk_2026_02_01_082540) |
+| ALE/Berzerk-v5 | 1072 | ppo_atari | [ppo_atari_berzerk_2026_01_06_112515](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_berzerk_2026_01_06_112515) |
+| ALE/Bowling-v5 | 23.96 | a2c_gae_atari | [a2c_gae_atari_bowling_2026_02_01_082529](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_bowling_2026_02_01_082529) |
+| ALE/Bowling-v5 | 46.45 | ppo_atari | [ppo_atari_bowling_2026_01_06_113148](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_bowling_2026_01_06_113148) |
+| ALE/Boxing-v5 | 1.80 | a2c_gae_atari | [a2c_gae_atari_boxing_2026_02_01_082539](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_boxing_2026_02_01_082539) |
+| ALE/Boxing-v5 | 91.17 | ppo_atari | [ppo_atari_boxing_2026_01_06_112531](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_boxing_2026_01_06_112531) |
+| ALE/Breakout-v5 | 273 | a2c_gae_atari | [a2c_gae_atari_breakout_2026_01_31_213610](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_breakout_2026_01_31_213610) |
+| ALE/Breakout-v5 | 327 | ppo_atari_lam70 | [ppo_atari_lam70_breakout_2026_01_07_110559](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_breakout_2026_01_07_110559) |
+| ALE/Carnival-v5 | 2170 | a2c_gae_atari | [a2c_gae_atari_carnival_2026_02_01_082726](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_carnival_2026_02_01_082726) |
+| ALE/Carnival-v5 | 3967 | ppo_atari_lam70 | [ppo_atari_lam70_carnival_2026_01_07_144738](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_carnival_2026_01_07_144738) |
+| ALE/Centipede-v5 | 1382 | a2c_gae_atari | [a2c_gae_atari_centipede_2026_02_01_082643](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_centipede_2026_02_01_082643) |
+| ALE/Centipede-v5 | 4915 | ppo_atari_lam70 | [ppo_atari_lam70_centipede_2026_01_07_223557](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_centipede_2026_01_07_223557) |
+| ALE/ChopperCommand-v5 | 2446 | a2c_gae_atari | [a2c_gae_atari_choppercommand_2026_02_01_082626](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_choppercommand_2026_02_01_082626) |
+| ALE/ChopperCommand-v5 | 5355 | ppo_atari | [ppo_atari_choppercommand_2026_01_07_110539](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_choppercommand_2026_01_07_110539) |
+| ALE/CrazyClimber-v5 | 96943 | a2c_gae_atari | [a2c_gae_atari_crazyclimber_2026_02_01_082625](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_crazyclimber_2026_02_01_082625) |
+| ALE/CrazyClimber-v5 | 107370 | ppo_atari_lam85 | [ppo_atari_lam85_crazyclimber_2026_01_07_223609](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_crazyclimber_2026_01_07_223609) |
+| ALE/Defender-v5 | 33149 | a2c_gae_atari | [a2c_gae_atari_defender_2026_02_01_082658](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_defender_2026_02_01_082658) |
+| ALE/Defender-v5 | 51439 | ppo_atari_lam70 | [ppo_atari_lam70_defender_2026_01_07_205238](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_defender_2026_01_07_205238) |
+| ALE/DemonAttack-v5 | 2962 | a2c_gae_atari | [a2c_gae_atari_demonattack_2026_02_01_082717](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_demonattack_2026_02_01_082717) |
+| ALE/DemonAttack-v5 | 16558 | ppo_atari_lam70 | [ppo_atari_lam70_demonattack_2026_01_07_111315](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_demonattack_2026_01_07_111315) |
+| ALE/DoubleDunk-v5 | -1.69 | a2c_gae_atari | [a2c_gae_atari_doubledunk_2026_02_01_082901](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_doubledunk_2026_02_01_082901) |
+| ALE/DoubleDunk-v5 | -2.38 | ppo_atari | [ppo_atari_doubledunk_2026_01_07_110802](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_doubledunk_2026_01_07_110802) |
+| ALE/ElevatorAction-v5 | 731 | a2c_gae_atari | [a2c_gae_atari_elevatoraction_2026_02_01_082908](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_elevatoraction_2026_02_01_082908) |
+| ALE/ElevatorAction-v5 | 5446 | ppo_atari | [ppo_atari_elevatoraction_2026_01_06_113129](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_elevatoraction_2026_01_06_113129) |
+| ALE/Enduro-v5 | 681 | a2c_gae_atari | [a2c_gae_atari_enduro_2026_02_01_001123](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_enduro_2026_02_01_001123) |
+| ALE/Enduro-v5 | 898 | ppo_atari_lam85 | [ppo_atari_lam85_enduro_2026_01_08_095448](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_enduro_2026_01_08_095448) |
+| ALE/FishingDerby-v5 | -16.38 | a2c_gae_atari | [a2c_gae_atari_fishingderby_2026_02_01_082906](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_fishingderby_2026_02_01_082906) |
+| ALE/FishingDerby-v5 | 27.10 | ppo_atari_lam85 | [ppo_atari_lam85_fishingderby_2026_01_08_094158](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_fishingderby_2026_01_08_094158) |
+| ALE/Freeway-v5 | 23.13 | a2c_gae_atari | [a2c_gae_atari_freeway_2026_02_01_082931](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_freeway_2026_02_01_082931) |
+| ALE/Freeway-v5 | 31.30 | ppo_atari | [ppo_atari_freeway_2026_01_06_182318](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_freeway_2026_01_06_182318) |
+| ALE/Frostbite-v5 | 266 | a2c_gae_atari | [a2c_gae_atari_frostbite_2026_02_01_082915](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_frostbite_2026_02_01_082915) |
+| ALE/Frostbite-v5 | 301 | ppo_atari | [ppo_atari_frostbite_2026_01_06_112556](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_frostbite_2026_01_06_112556) |
+| ALE/Gopher-v5 | 984 | a2c_gae_atari | [a2c_gae_atari_gopher_2026_02_01_133323](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_gopher_2026_02_01_133323) |
+| ALE/Gopher-v5 | 6508 | ppo_atari_lam70 | [ppo_atari_lam70_gopher_2026_01_07_170451](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_gopher_2026_01_07_170451) |
+| ALE/Gravitar-v5 | 270 | a2c_gae_atari | [a2c_gae_atari_gravitar_2026_02_01_133244](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_gravitar_2026_02_01_133244) |
+| ALE/Gravitar-v5 | 599 | ppo_atari | [ppo_atari_gravitar_2026_01_06_112548](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_gravitar_2026_01_06_112548) |
+| ALE/Hero-v5 | 18680 | a2c_gae_atari | [a2c_gae_atari_hero_2026_02_01_175903](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_hero_2026_02_01_175903) |
+| ALE/Hero-v5 | 28238 | ppo_atari_lam85 | [ppo_atari_lam85_hero_2026_01_07_223619](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_hero_2026_01_07_223619) |
+| ALE/IceHockey-v5 | -5.92 | a2c_gae_atari | [a2c_gae_atari_icehockey_2026_02_01_175745](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_icehockey_2026_02_01_175745) |
+| ALE/IceHockey-v5 | -3.93 | ppo_atari | [ppo_atari_icehockey_2026_01_06_183721](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_icehockey_2026_01_06_183721) |
+| ALE/Jamesbond-v5 | 460 | a2c_gae_atari | [a2c_gae_atari_jamesbond_2026_02_01_175945](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_jamesbond_2026_02_01_175945) |
+| ALE/Jamesbond-v5 | 662 | ppo_atari | [ppo_atari_jamesbond_2026_01_06_183717](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_jamesbond_2026_01_06_183717) |
+| ALE/JourneyEscape-v5 | -965 | a2c_gae_atari | [a2c_gae_atari_journeyescape_2026_02_01_084415](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_journeyescape_2026_02_01_084415) |
+| ALE/JourneyEscape-v5 | -1252 | ppo_atari_lam85 | [ppo_atari_lam85_journeyescape_2026_01_08_094842](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_journeyescape_2026_01_08_094842) |
+| ALE/Kangaroo-v5 | 322 | a2c_gae_atari | [a2c_gae_atari_kangaroo_2026_02_01_084415](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_kangaroo_2026_02_01_084415) |
+| ALE/Kangaroo-v5 | 9912 | ppo_atari_lam85 | [ppo_atari_lam85_kangaroo_2026_01_07_110838](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_kangaroo_2026_01_07_110838) |
+| ALE/Krull-v5 | 7519 | a2c_gae_atari | [a2c_gae_atari_krull_2026_02_01_084420](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_krull_2026_02_01_084420) |
+| ALE/Krull-v5 | 7841 | ppo_atari | [ppo_atari_krull_2026_01_07_110747](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_krull_2026_01_07_110747) |
+| ALE/KungFuMaster-v5 | 23006 | a2c_gae_atari | [a2c_gae_atari_kungfumaster_2026_02_01_085101](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_kungfumaster_2026_02_01_085101) |
+| ALE/KungFuMaster-v5 | 29068 | ppo_atari_lam70 | [ppo_atari_lam70_kungfumaster_2026_01_07_111317](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_kungfumaster_2026_01_07_111317) |
+| ALE/MsPacman-v5 | 2110 | a2c_gae_atari | [a2c_gae_atari_mspacman_2026_02_01_001100](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_mspacman_2026_02_01_001100) |
+| ALE/MsPacman-v5 | 2372 | ppo_atari_lam85 | [ppo_atari_lam85_mspacman_2026_01_07_223522](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_mspacman_2026_01_07_223522) |
+| ALE/NameThisGame-v5 | 5412 | a2c_gae_atari | [a2c_gae_atari_namethisgame_2026_02_01_132733](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_namethisgame_2026_02_01_132733) |
+| ALE/NameThisGame-v5 | 5993 | ppo_atari | [ppo_atari_namethisgame_2026_01_06_182952](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_namethisgame_2026_01_06_182952) |
+| ALE/Phoenix-v5 | 5635 | a2c_gae_atari | [a2c_gae_atari_phoenix_2026_02_01_085101](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_phoenix_2026_02_01_085101) |
+| ALE/Phoenix-v5 | 15659 | ppo_atari_lam70 | [ppo_atari_lam70_phoenix_2026_01_07_110832](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_phoenix_2026_01_07_110832) |
+| ALE/Pong-v5 | 10.17 | a2c_gae_atari | [a2c_gae_atari_pong_2026_01_31_213635](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_pong_2026_01_31_213635) |
+| ALE/Pong-v5 | 16.91 | ppo_atari_lam85 | [ppo_atari_lam85_pong_2026_01_08_094454](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_pong_2026_01_08_094454) |
+| ALE/Pooyan-v5 | 2997 | a2c_gae_atari | [a2c_gae_atari_pooyan_2026_02_01_132748](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_pooyan_2026_02_01_132748) |
+| ALE/Pooyan-v5 | 5716 | ppo_atari_lam70 | [ppo_atari_lam70_pooyan_2026_01_07_224346](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_pooyan_2026_01_07_224346) |
+| ALE/Qbert-v5 | 12619 | a2c_gae_atari | [a2c_gae_atari_qbert_2026_01_31_213720](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_qbert_2026_01_31_213720) |
+| ALE/Qbert-v5 | 15094 | ppo_atari | [ppo_atari_qbert_2026_01_06_111801](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_qbert_2026_01_06_111801) |
+| ALE/Riverraid-v5 | 6558 | a2c_gae_atari | [a2c_gae_atari_riverraid_2026_02_01_132507](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_riverraid_2026_02_01_132507) |
+| ALE/Riverraid-v5 | 9428 | ppo_atari_lam85 | [ppo_atari_lam85_riverraid_2026_01_07_204356](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_riverraid_2026_01_07_204356) |
+| ALE/RoadRunner-v5 | 29810 | a2c_gae_atari | [a2c_gae_atari_roadrunner_2026_02_01_132509](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_roadrunner_2026_02_01_132509) |
+| ALE/RoadRunner-v5 | 37015 | ppo_atari_lam85 | [ppo_atari_lam85_roadrunner_2026_01_07_145913](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_roadrunner_2026_01_07_145913) |
+| ALE/Robotank-v5 | 2.80 | a2c_gae_atari | [a2c_gae_atari_robotank_2026_02_01_132434](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_robotank_2026_02_01_132434) |
+| ALE/Robotank-v5 | 20.07 | ppo_atari | [ppo_atari_robotank_2026_01_06_183413](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_robotank_2026_01_06_183413) |
+| ALE/Seaquest-v5 | 850 | a2c_gae_atari | [a2c_gae_atari_seaquest_2026_02_01_001001](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_seaquest_2026_02_01_001001) |
+| ALE/Seaquest-v5 | 1796 | ppo_atari | [ppo_atari_seaquest_2026_01_06_183440](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_seaquest_2026_01_06_183440) |
+| ALE/Skiing-v5 | -14235 | a2c_gae_atari | [a2c_gae_atari_skiing_2026_02_01_132451](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_skiing_2026_02_01_132451) |
+| ALE/Skiing-v5 | -19340 | ppo_atari | [ppo_atari_skiing_2026_01_06_183424](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_skiing_2026_01_06_183424) |
+| ALE/Solaris-v5 | - | a2c_gae_atari | - |
+| ALE/Solaris-v5 | 2094 | ppo_atari | [ppo_atari_solaris_2026_01_06_192643](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_solaris_2026_01_06_192643) |
+| ALE/SpaceInvaders-v5 | 784 | a2c_gae_atari | [a2c_gae_atari_spaceinvaders_2026_02_01_000950](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_spaceinvaders_2026_02_01_000950) |
+| ALE/SpaceInvaders-v5 | 726 | ppo_atari | [ppo_atari_spaceinvaders_2026_01_07_102346](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_spaceinvaders_2026_01_07_102346) |
+| ALE/StarGunner-v5 | 8665 | a2c_gae_atari | [a2c_gae_atari_stargunner_2026_02_01_132406](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_stargunner_2026_02_01_132406) |
+| ALE/StarGunner-v5 | 47495 | ppo_atari_lam70 | [ppo_atari_lam70_stargunner_2026_01_07_111404](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_stargunner_2026_01_07_111404) |
+| ALE/Surround-v5 | -9.72 | a2c_gae_atari | [a2c_gae_atari_surround_2026_02_01_132215](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_surround_2026_02_01_132215) |
+| ALE/Surround-v5 | -2.52 | ppo_atari | [ppo_atari_surround_2026_01_07_102404](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_surround_2026_01_07_102404) |
+| ALE/Tennis-v5 | -2873 | a2c_gae_atari | [a2c_gae_atari_tennis_2026_02_01_175829](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_tennis_2026_02_01_175829) |
+| ALE/Tennis-v5 | -4.41 | ppo_atari_lam85 | [ppo_atari_lam85_tennis_2026_01_07_223532](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_tennis_2026_01_07_223532) |
+| ALE/TimePilot-v5 | 3376 | a2c_gae_atari | [a2c_gae_atari_timepilot_2026_02_01_175930](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_timepilot_2026_02_01_175930) |
+| ALE/TimePilot-v5 | 4668 | ppo_atari | [ppo_atari_timepilot_2026_01_07_101010](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_timepilot_2026_01_07_101010) |
+| ALE/Tutankham-v5 | 167 | a2c_gae_atari | [a2c_gae_atari_tutankham_2026_02_01_132347](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_tutankham_2026_02_01_132347) |
+| ALE/Tutankham-v5 | 217 | ppo_atari_lam85 | [ppo_atari_lam85_tutankham_2026_01_08_095251](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam85_tutankham_2026_01_08_095251) |
+| ALE/UpNDown-v5 | 57099 | a2c_gae_atari | [a2c_gae_atari_upndown_2026_02_01_132435](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_upndown_2026_02_01_132435) |
+| ALE/UpNDown-v5 | 182472 | ppo_atari | [ppo_atari_upndown_2026_01_07_105708](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_upndown_2026_01_07_105708) |
+| ALE/VideoPinball-v5 | 25310 | a2c_gae_atari | [a2c_gae_atari_videopinball_2026_02_01_083457](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_videopinball_2026_02_01_083457) |
+| ALE/VideoPinball-v5 | 56746 | ppo_atari_lam70 | [ppo_atari_lam70_videopinball_2026_01_07_224359](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_lam70_videopinball_2026_01_07_224359) |
+| ALE/WizardOfWor-v5 | 2682 | a2c_gae_atari | [a2c_gae_atari_wizardofwor_2026_02_01_132449](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_wizardofwor_2026_02_01_132449) |
+| ALE/WizardOfWor-v5 | 5814 | ppo_atari | [ppo_atari_wizardofwor_2026_01_06_221154](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_wizardofwor_2026_01_06_221154) |
+| ALE/YarsRevenge-v5 | 24371 | a2c_gae_atari | [a2c_gae_atari_yarsrevenge_2026_02_01_132224](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_yarsrevenge_2026_02_01_132224) |
+| ALE/YarsRevenge-v5 | 17120 | ppo_atari | [ppo_atari_yarsrevenge_2026_01_06_221154](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_yarsrevenge_2026_01_06_221154) |
+| ALE/Zaxxon-v5 | 29.46 | a2c_gae_atari | [a2c_gae_atari_zaxxon_2026_02_01_131758](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/a2c_gae_atari_zaxxon_2026_02_01_131758) |
+| ALE/Zaxxon-v5 | 10756 | ppo_atari | [ppo_atari_zaxxon_2026_01_06_221154](https://huggingface.co/datasets/SLM-Lab/benchmark/tree/main/data/ppo_atari_zaxxon_2026_01_06_221154) |
+
+**Skipped** (hard exploration): Adventure, MontezumaRevenge, Pitfall, PrivateEye, Venture
+
+Full Lambda Comparison Table (click to expand)
+
+| ENV | ppo_atari | ppo_atari_lam85 | ppo_atari_lam70 |
+|-----|-----------|-----------------|-----------------|
| ALE/AirRaid-v5 | **8245** | - | - |
| ALE/Alien-v5 | **1453** | 1353 | 1274 |
| ALE/Amidar-v5 | 574 | **580** | - |
@@ -443,14 +572,11 @@ source .env && slm-lab run-remote --gpu -s env=ENV \
| ALE/Kangaroo-v5 | 2623 | **9912** | - |
| ALE/Krull-v5 | **7841** | - | - |
| ALE/KungFuMaster-v5 | 18973 | 28334 | **29068** |
-| ALE/MontezumaRevenge-v5 | Skip | Skip | Skip |
| ALE/MsPacman-v5 | 2308 | **2372** | 2297 |
| ALE/NameThisGame-v5 | **5993** | - | - |
| ALE/Phoenix-v5 | 7940 | - | **15659** |
-| ALE/Pitfall-v5 | Skip | Skip | Skip |
| ALE/Pong-v5 | 15.01 | **16.91** | 12.85 |
| ALE/Pooyan-v5 | 4704 | - | **5716** |
-| ALE/PrivateEye-v5 | Skip | Skip | Skip |
| ALE/Qbert-v5 | **15094** | - | - |
| ALE/Riverraid-v5 | 7319 | **9428** | - |
| ALE/RoadRunner-v5 | 24204 | **37015** | - |
@@ -465,40 +591,114 @@ source .env && slm-lab run-remote --gpu -s env=ENV \
| ALE/TimePilot-v5 | **4668** | - | - |
| ALE/Tutankham-v5 | 203 | **217** | - |
| ALE/UpNDown-v5 | **182472** | - | - |
-| ALE/Venture-v5 | Skip | Skip | Skip |
| ALE/VideoPinball-v5 | 31385 | - | **56746** |
| ALE/WizardOfWor-v5 | **5814** | 5466 | 4740 |
| ALE/YarsRevenge-v5 | **17120** | - | - |
| ALE/Zaxxon-v5 | **10756** | - | - |
-**Legend**: **Bold** = Best score | Skip = Hard exploration | - = Not tested
-
----
-
-#### Sticky Actions Validation (v5 vs v4-style)
-
-Testing hypothesis that lower scores are due to sticky actions (`repeat_action_probability=0.25` in v5 vs `0.0` in v4/CleanRL).
-
-**Environment**: Same as above, but with `repeat_action_probability=0.0` (matching CleanRL/old v4 behavior)
-
-**Reproduce**: Copy `ENV` from first column:
-```bash
-source .env && slm-lab run-remote --gpu -s env=ENV \
- slm_lab/spec/benchmark/ppo/ppo_atari.json ppo_atari_nosticky train -n NAME
-```
-
-**Results** (Testing games with significant regression):
-
-| ENV | v5 (sticky=0.25) | v4-style (sticky=0.0) | Diff | % Change |
-| --- | ---------------- | --------------------- | ---- | -------- |
-| ALE/Skiing-v5 | -19340 | - | - | - |
-| ALE/Frostbite-v5 | 301 | - | - | - |
-| ALE/ElevatorAction-v5 | 5446 | - | - | - |
-| ALE/Gravitar-v5 | 599 | - | - | - |
-| ALE/WizardOfWor-v5 | 5814 | - | - | - |
-| ALE/Alien-v5 | 1453 | - | - | - |
-| ALE/KungFuMaster-v5 | 29068 | - | - | - |
-| ALE/Atlantis-v5 | 792886 | - | - | - |
-| ALE/Pong-v5 | 15.01 | - | - | - |
-| ALE/Breakout-v5 | 191 | - | - | - |
-
+**Legend**: **Bold** = Best score | - = Not tested
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+