codelion commited on
Commit
8ea3ca1
Β·
verified Β·
1 Parent(s): 238fadc

Fix rankings: #1 Optimal Accuracy (88.7%), #2 Optimal Selection (43.0%), #7 overall / #5 open-source

Browse files
Files changed (1) hide show
  1. README.md +25 -15
README.md CHANGED
@@ -19,15 +19,18 @@ metrics:
19
 
20
  ## Performance
21
 
22
- πŸ† **#1 on RouterArena Leaderboard in Optimal Accuracy Score**
23
 
24
  Official RouterArena Full Dataset Results (8,400 queries):
25
- - **88.7% Optimal Accuracy Score** - πŸ₯‡ SOTA! Ranked #1 in this category
26
- - **64.9% Overall Accuracy** - #1 among open-source routers
27
- - **Arena Score: 63.8**
 
28
  - **$0.60 per 1K queries** - Cost-efficient routing
29
 
30
- The **Optimal Accuracy Score** measures how often the router makes the right routing decision - when Chayan selects a model for a query, that model provides the correct answer 88.7% of the time.
 
 
31
 
32
  Sub_10 Benchmark (809 queries):
33
  - **69.05% accuracy**
@@ -164,22 +167,29 @@ predictions = router.predict(augmented, k=4)
164
 
165
  ## RouterArena Leaderboard
166
 
167
- πŸ† **Official Results - #1 in Optimal Accuracy Score Category**
168
 
169
  ![RouterArena Leaderboard](routerarena_leaderboard.png)
170
 
171
- Chayan on the official [RouterArena leaderboard](https://routeworks.github.io/):
172
 
173
- | Rank (Overall) | Router | Arena Score | Accuracy | **Opt. Acc** | Cost/1k | Type |
174
- |----------------|--------|-------------|----------|--------------|---------|------|
175
- | 1 | **Chayan** | 63.8 | 64.9% | **88.7%** πŸ₯‡ | $0.60 | Open-Source |
176
- | 2 | RouterBench-MLP | 57.6 | 61.6% | 83.3% | $4.80 | Open-Source |
177
- | 3 | Azure | 66.7 | 68.1% | 82.0% | $0.50 | Closed-Source |
178
- | 4 | vLLM-SR | 64.3 | 67.3% | 79.3% | $1.70 | Open-Source |
 
179
 
180
- **πŸ₯‡ SOTA Achievement - Optimal Accuracy Score Category**: Chayan achieves **88.7% Optimal Accuracy**, ranking **#1** in this critical metric across all routers on the leaderboard.
 
 
 
 
181
 
182
- **What is Optimal Accuracy Score?** This metric measures routing decision quality - when Chayan selects a model for a query, that model provides the correct answer 88.7% of the time. This is the highest score among all evaluated routers, demonstrating Chayan's superior model selection capability.
 
 
183
 
184
  View the full leaderboard and PR: [RouterArena PR #24](https://github.com/RouteWorks/RouterArena/pull/24)
185
 
 
19
 
20
  ## Performance
21
 
22
+ πŸ† **SOTA on RouterArena: #1 Optimal Accuracy Score, #2 Optimal Selection**
23
 
24
  Official RouterArena Full Dataset Results (8,400 queries):
25
+ - **88.7% Optimal Accuracy Score** - πŸ₯‡ #1 SOTA! (Best routing decision quality)
26
+ - **43.0% Optimal Selection Score** - πŸ₯ˆ #2 Silver! (Second-best model selection)
27
+ - **64.9% Overall Accuracy** - #7 overall, #5 among open-source routers
28
+ - **Arena Score: 63.8** - #7 overall, #5 among open-source routers
29
  - **$0.60 per 1K queries** - Cost-efficient routing
30
 
31
+ **Key Metrics Explained:**
32
+ - **Optimal Accuracy Score**: When Chayan routes to a model, that model gives the correct answer 88.7% of the time (highest on leaderboard!)
33
+ - **Optimal Selection Score**: Chayan selects the best available model 43% of the time (second-best on leaderboard)
34
 
35
  Sub_10 Benchmark (809 queries):
36
  - **69.05% accuracy**
 
167
 
168
  ## RouterArena Leaderboard
169
 
170
+ πŸ† **Official Results - #1 Optimal Accuracy, #2 Optimal Selection**
171
 
172
  ![RouterArena Leaderboard](routerarena_leaderboard.png)
173
 
174
+ Chayan on the official [RouterArena leaderboard](https://routeworks.github.io/) (sorted by Optimal Accuracy):
175
 
176
+ | Rank | Router | **Opt. Acc** | **Opt. Select** | Accuracy | Arena Score | Cost/1k | Type |
177
+ |------|--------|--------------|-----------------|----------|-------------|---------|------|
178
+ | **1** | **Chayan** | **88.7%** πŸ₯‡ | **43.0%** πŸ₯ˆ | 64.9% | 63.8 | $0.60 | Open-Source |
179
+ | 2 | RouterBench-MLP | 83.3% | 13.4% | 61.6% | 57.6 | $4.80 | Open-Source |
180
+ | 3 | Azure | 82.0% | 22.5% | 68.1% | 66.7 | $0.50 | Closed-Source |
181
+ | 4 | vLLM-SR | 79.3% | 4.8% | 67.3% | 64.3 | $1.70 | Open-Source |
182
+ | 5 | RouterBench-KNN | 78.8% | 13.1% | 58.7% | 55.5 | $4.30 | Open-Source |
183
 
184
+ **Overall Rankings:**
185
+ - **Optimal Accuracy Score**: πŸ₯‡ #1 (88.7% - SOTA!)
186
+ - **Optimal Selection Score**: πŸ₯ˆ #2 (43.0% - Silver!)
187
+ - **Overall Accuracy**: #7 overall, #5 among open-source
188
+ - **Arena Score**: #7 overall, #5 among open-source
189
 
190
+ **πŸ₯‡ SOTA Achievement - Optimal Accuracy Score**: Chayan achieves **88.7%**, the highest score among all routers. This measures routing decision quality - when Chayan routes to a model, that model gives the correct answer 88.7% of the time.
191
+
192
+ **πŸ₯ˆ Silver Medal - Optimal Selection Score**: Chayan ranks **#2** with **43.0%**, meaning it selects the optimal model (best among all available models) 43% of the time, second only to Azure's 46.3%.
193
 
194
  View the full leaderboard and PR: [RouterArena PR #24](https://github.com/RouteWorks/RouterArena/pull/24)
195