Model Comparison
Qwen3 235B A22B (Non-reasoning)
vs. Reka Flash (Sep '24)
Comparing 2 AI models · 12 benchmarks · Alibaba, Reka AI
Recommended Pick
AlQwen3 235B A22B (Non-reasoning)5 metric winsStrongest on: TTFT, Latency, Reasoning
Best Value
Reka Flash (Sep '24)
100.0 value score
29.6 reasoning / $0.35/1M
Lowest Price
Reka Flash (Sep '24)
$0.20/1M input price
Best Reasoning
Qwen3 235B A22B (Non-reasoning)
35.3 reasoning score
Blends available reasoning benchmarks
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
Reka Flash (Sep '24) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Reka Flash (Sep '24) is 3.5x cheaper on input tokens than Qwen3 235B A22B (Non-reasoning).
Speed gap
Reka Flash (Sep '24) generates about 1.2x as many tokens per second as Qwen3 235B A22B (Non-reasoning).
Reasoning gap
Qwen3 235B A22B (Non-reasoning) leads Reka Flash (Sep '24) by 5.7 points on reasoning.
Top-pick rationale
Qwen3 235B A22B (Non-reasoning) wins 5 measurable categories, including TTFT, Latency, Reasoning, Intelligence.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Qwen3 235B A22B (Non-reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Reka Flash (Sep '24)
Reka AI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Chat with leading AI models
Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, Qwen & Kimi.
EU-hosted inference
Servers in Germany & Finland. Designed to meet strict GDPR and ISO 27001 compliance requirements.
Full Comparison
| Metric | Top Pick Al Qwen3 235B A22B (Non-reasoning) | Re Reka Flash (Sep '24) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.70/1M | $0.20/1M |
| Output Cost | $2.80/1M | $0.80/1M |
| Blended (3:1) | $1.22/1M | $0.35/1M |
| Specifications | ||
| Organization | Alibaba | Reka AI |
| Release Date | Apr 28, 2025 | Oct 4, 2024 |
| Performance & Speed | ||
| Throughput | 68.6 tok/s | 85.4 tok/s |
| TTFT | 1114ms | 1725ms |
| Latency | 1114ms | 1725ms |
| Composite Indices | ||
| Value Score | 34.1 | 100.0 |
| Reasoning Score | 35.3 | 29.6 |
| Intelligence | 10.9 | 6.3 |
| Math | 23.7 | — |
| Standard Benchmarks | ||
| GPQA | 61.3% | — |
| MMLU Pro | 76.2% | — |
| HLE | 4.7% | — |
| LiveCodeBench | 34.3% | — |
| MATH 500 | 90.2% | 52.9% |
| AIME 2025 | 23.7% | — |
| AIME (Original) | 32.7% | — |
| SciCode | 29.9% | — |
| LCR | 0.0% | — |
| IFBench | 36.6% | — |
| TAU-bench v2 | 27.2% | — |
| TerminalBench Hard | 6.1% | — |
Key Takeaways
Reka Flash (Sep '24) offers the best value at $0.20/1M,making it ideal for high-volume applications and cost-conscious projects.
Qwen3 235B A22B (Non-reasoning) has the strongest reasoning profile with a 35.3 reasoning score,combining the available reasoning-heavy benchmarks.
All models support context windows of ∞+ tokens,suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Qwen3 235B A22B (Non-reasoning)
- Complex reasoning tasks
- Research & analysis
Reka Flash (Sep '24)
- Cost-sensitive applications
- High-volume processing
