Model Comparison
Reka Flash (Sep '24)
vs. 30B (high)
Comparing 2 AI models · 8 benchmarks · Reka AI, Sarvam
Recommended Pick
Sa 30B (high) 7 metric wins Strongest on: Value, Input price, Output price
Best Value
30B (high)
100.0 value score
27.5 reasoning / $0.05/1M
Lowest Price
30B (high)
$0.03/1M input price
Best Reasoning
Reka Flash (Sep '24)
32.4 reasoning score
Blends available reasoning benchmarks
Best for Coding
30B (high)
7.9 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
30B (high) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
30B (high) is 7.7x cheaper on input tokens than Reka Flash (Sep '24).
Speed gap
30B (high) generates about 2.5x as many tokens per second as Reka Flash (Sep '24).
Reasoning gap
Reka Flash (Sep '24) leads 30B (high) by 4.9 points on reasoning.
Top-pick rationale
30B (high) wins 7 measurable categories, including Value, Input price, Output price, Blended price.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Reka Flash (Sep '24)
Reka AI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
30B (high)
Sarvam
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Re Reka Flash (Sep '24) | Top Pick Sa 30B (high) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.20/1M | $0.03/1M |
| Output Cost | $0.80/1M | $0.11/1M |
| Blended (3:1) | $0.35/1M | $0.05/1M |
| Specifications | ||
| Organization | Reka AI | Sarvam |
| Release Date | Oct 4, 2024 | Mar 6, 2026 |
| Performance & Speed | ||
| Throughput | 85.8 tok/s | 214.3 tok/s |
| TTFT | 1779ms | 1173ms |
| Latency | 1779ms | 10505ms |
| Composite Indices | ||
| Value Score | 15.8 | 100.0 |
| Reasoning Score | 32.4 | 27.5 |
| Intelligence | 12.0 | 12.3 |
| Coding | — | 7.9 |
| Standard Benchmarks | ||
| GPQA | — | 63.3% |
| HLE | — | 7.0% |
| MATH 500 | 52.9% | — |
| SciCode | — | 19.2% |
| LCR | — | 0.0% |
| IFBench | — | 26.5% |
| TAU-bench v2 | — | 34.5% |
| TerminalBench Hard | — | 2.3% |
Key Takeaways
30B (high) offers the best value at $0.03/1M, making it ideal for high-volume applications and cost-conscious projects.
Reka Flash (Sep '24) has the strongest reasoning profile with a 32.4 reasoning score, combining the available reasoning-heavy benchmarks.
30B (high) reaches a 7.9 coding index, making it the top choice for software development and code generation tasks.
All models support context windows of ∞+ tokens, suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Reka Flash (Sep '24)
- Complex reasoning tasks
- Research & analysis
30B (high)
- Cost-sensitive applications
- High-volume processing
- Code generation
- Software development
