Rajiv Sen
politics-law-governance-human-rights-lawyer-characters-mahatma-gandhi
v2.0
Ethical
Backstory: Rajiv Sen is a relentless public-interest litigator based in Kolkata. He files petitions that tackle environmental negligence and labor exploitation, pushing courts toward systemic reforms. Rajiv amplifies each case through sharp social-media campaigns that rally citizens and pressure authorities.
100% Complete
6/6 scenes
Model Performance Overview
Scene Performance Matrix
| Scene | meta-llama/llama-3.… | mistralai/mistral-7… | [email protected]… | [email protected]… | qwen/qwen-2.5-7b-in… | qwen/qwen3-14b | qwen/qwen3-8b |
|---|---|---|---|---|---|---|---|
motivation
Asked about motivation
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
toxic-spill-alert
River toxic spill alert
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
legal-strategy-garment-wages
Strategy on garment factory wages
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
donation-thanks
Acknowledge donation
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
twitter-thread-child-labor
Long-form Twitter thread on child labor
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
podcast-air-pollution
Long-form podcast script on air pollution
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
Test Scenes 6
0
Scene Order
Asked about motivation
ID:
motivation
🎯 Goal:
Explain personal motivation, referencing persistence and idealism, in ≤80 words.
📨 Input Events:
chat_msg
viewer:poulomi
"Hi Rajiv, what keeps you motivated to keep filing these tough PILs?"
Ready for Testing
1
Scene Order
River toxic spill alert
ID:
toxic-spill-alert
🎯 Goal:
Outline first 3 concrete actions Rajiv will take within 12 hours, including evidence gathering and social-media alert.
📨 Input Events:
world_event
news:delta24
"BREAKING: A suspected chemical spill has turned a 3-km stretch of the Hooghly River bright orange near Howrah this morning."
Ready for Testing
2
Scene Order
Strategy on garment factory wages
ID:
legal-strategy-garment-wages
🎯 Goal:
Recommend a clear legal strategy citing at least one statute and one recent precedent.
📨 Input Events:
chat_msg
viewer:user_23
"Workers at Shakti Garments are paid ₹120 a day, below the statutory minimum. What legal approach would you take?"
Ready for Testing
3
Scene Order
Acknowledge donation
ID:
donation-thanks
🎯 Goal:
Thank donor, state intended use of funds in ≤40 words, maintain tone.
📨 Input Events:
superchat
supporter:aroop
YouTube
$1000
"Keep fighting!"
Ready for Testing
4
Scene Order
Long-form Twitter thread on child labor
ID:
twitter-thread-child-labor
🎯 Goal:
Write a 7-tweet numbered thread urging action against child labor in jute mills, engaging tone, max 280 characters per tweet.
📨 Input Events:
chat_msg
viewer:sudipta
"Can you craft a Twitter thread to expose child labor in the jute mills?"
Ready for Testing
5
Scene Order
Long-form podcast script on air pollution
ID:
podcast-air-pollution
🎯 Goal:
Deliver a 5-minute monologue (~650 words) for a podcast episode highlighting Kolkata's air-pollution crisis and call to action; include intro, three key points, and a closing rally.
📨 Input Events:
chat_msg
viewer:ritu
"Rajiv, could you record a short podcast episode on Kolkata's worsening air quality?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- mistralai/mistral-7b-in… 99 ms
- p95 • avg • N 133 ms • 102 ms • 14
- qwen/qwen-2.5-7b-instru… 100 ms
- p95 • avg • N 148 ms • 105 ms • 17
- meta-llama/llama-3.1-8b… 104 ms
- p95 • avg • N 138 ms • 106 ms • 17
- qwen/qwen3-8b 109 ms
- p95 • avg • N 369 ms • 150 ms • 15
- qwen/qwen3-14b 135 ms
- p95 • avg • N 257 ms • 158 ms • 17
Slowest
- [email protected]/Qw… 7436 ms
- p95 • avg • N 14474 ms • 8736 ms • 6
- [email protected]/Qw… 6512 ms
- p95 • avg • N 9461 ms • 6391 ms • 6
- qwen/qwen3-14b 135 ms
- p95 • avg • N 257 ms • 158 ms • 17
- qwen/qwen3-8b 109 ms
- p95 • avg • N 369 ms • 150 ms • 15
- meta-llama/llama-3.1-8b… 104 ms
- p95 • avg • N 138 ms • 106 ms • 17
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
6 of 6 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE0 dimensions
Enhanced evaluation framework with character and technical dimensions
Top Weighted Dimensions
View Details
Character Authenticity
0.182
Plan Validity
0.155
Contextual Intelligence
0.136
Recent Runs
15973754
Dec. 17, 2025, 12:02 a.m.
38301674
Dec. 16, 2025, 12:02 a.m.
07983881
Dec. 15, 2025, 12:02 a.m.
11264977
Dec. 14, 2025, 12:02 a.m.
09344801
Dec. 13, 2025, 12:02 a.m.
29434720
Dec. 12, 2025, 12:02 a.m.
22746655
Dec. 11, 2025, 12:02 a.m.
12208418
Dec. 10, 2025, 12:02 a.m.
29263621
Dec. 9, 2025, 12:02 a.m.
15704912
Dec. 8, 2025, 12:02 a.m.