Model Comparison
LFM2 8B A1B
vs. Qwen3 4B 2507 (Reasoning)
Comparing 2 AI models · 10 benchmarks · Liquid AI, Alibaba
Recommended Pick
Al Qwen3 4B 2507 (Reasoning) 14 metric wins Strongest on: Reasoning, Intelligence, Coding
Lowest Price
LFM2 8B A1B
$0.00/1M input price
Best Reasoning
Qwen3 4B 2507 (Reasoning)
51.2 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3 4B 2507 (Reasoning)
9.5 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Reasoning gap
Qwen3 4B 2507 (Reasoning) leads LFM2 8B A1B by 31.8 points on reasoning.
Coding gap
Qwen3 4B 2507 (Reasoning) leads LFM2 8B A1B by 7.2 points on coding.
Top-pick rationale
Qwen3 4B 2507 (Reasoning) wins 14 measurable categories, including Reasoning, Intelligence, Coding, Math.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
LFM2 8B A1B
Liquid AI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Qwen3 4B 2507 (Reasoning)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Li LFM2 8B A1B | Top Pick Al Qwen3 4B 2507 (Reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.00/1M | $0.00/1M |
| Output Cost | $0.00/1M | $0.00/1M |
| Specifications | ||
| Organization | Liquid AI | Alibaba |
| Release Date | Oct 7, 2025 | Aug 6, 2025 |
| Performance & Speed | ||
| Throughput | — | — |
| TTFT | — | — |
| Latency | — | — |
| Composite Indices | ||
| Reasoning Score | 19.4 | 51.2 |
| Intelligence | 7.0 | 18.2 |
| Coding | 2.3 | 9.5 |
| Math | 25.3 | 82.7 |
| Standard Benchmarks | ||
| GPQA | 34.4% | 66.7% |
| MMLU Pro | 50.5% | 74.3% |
| HLE | 4.9% | 5.9% |
| LiveCodeBench | 15.1% | 64.1% |
| AIME 2025 | 25.3% | 82.7% |
| SciCode | 6.8% | 25.6% |
| LCR | 0.0% | 37.7% |
| IFBench | 26.3% | 49.8% |
| TAU-bench v2 | 10.5% | 25.4% |
| TerminalBench Hard | 0.0% | 1.5% |
Key Takeaways
LFM2 8B A1B offers the best value at $0.00/1M, making it ideal for high-volume applications and cost-conscious projects.
Qwen3 4B 2507 (Reasoning) has the strongest reasoning profile with a 51.2 reasoning score, combining the available reasoning-heavy benchmarks.
Qwen3 4B 2507 (Reasoning) reaches a 9.5 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
LFM2 8B A1B
- Cost-sensitive applications
- High-volume processing
Qwen3 4B 2507 (Reasoning)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
