Compare/Sarvam 105B (high) vs Qwen3.5 4B (Non-reasoning)

Sarvam 105B (high)vsQwen3.5 4B (Non-reasoning)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Sarvam

Sarvam 105B (high)

Input
$0.042/M
Output
$0.17/M
Speed
155 tok/s
TTFT
1.20s
Alibaba

Qwen3.5 4B (Non-reasoning)

Input
$0.03/M
Output
$0.15/M
Speed
28 tok/s
TTFT
0.42s

Winner by Category

Cheaper
Qwen3.5 4B (Non-reasoning)
Faster (tok/s)
Sarvam 105B (high)
Lower Latency
Qwen3.5 4B (Non-reasoning)
Benchmarks (4-3)
Sarvam 105B (high)

Pricing Comparison

MetricSarvam 105B (high)Qwen3.5 4B (Non-reasoning)
Input ($/M tokens)$0.042$0.03
Output ($/M tokens)$0.17$0.15
Cost for 1M input + 100K output tokens:
Sarvam 105B (high)$0.06
Qwen3.5 4B (Non-reasoning)$0.04

Speed Comparison

Output Speed (tokens/s) — higher is better
Sarvam 105B (high)
155 tok/s
Qwen3.5 4B (Non-reasoning)
28 tok/s
Time to First Token (seconds) — lower is better
Sarvam 105B (high)
1.20s
Qwen3.5 4B (Non-reasoning)
0.42s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
11.916.0
Coding Index
20.3
Math Index
GPQA Diamond
73.8%71.2%
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
10.1%7.5%
SciCode
26.4%18.3%
IFBench
34.4%33.3%
TerminalBench
1.5%11.4%
Sarvam 105B (high)4 wins
3 winsQwen3.5 4B (Non-reasoning)

Frequently Asked Questions

Which is cheaper, Sarvam 105B (high) or Qwen3.5 4B (Non-reasoning)?

Qwen3.5 4B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.06/M tokens vs $0.07/M for Sarvam 105B (high).

Which model performs better on benchmarks?

Sarvam 105B (high) wins 4 out of 12 benchmarks compared to 3 for Qwen3.5 4B (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Sarvam 105B (high) generates tokens faster at 155 tok/s vs 28 tok/s. However, Qwen3.5 4B (Non-reasoning) has lower time-to-first-token (0.42s vs 1.20s).

When should I use Sarvam 105B (high) vs Qwen3.5 4B (Non-reasoning)?

Choose based on your priorities: Qwen3.5 4B (Non-reasoning) for lower cost, Sarvam 105B (high) for stronger benchmark performance, and Sarvam 105B (high) for faster generation. For latency-sensitive apps, check the TTFT comparison above.