Qwen vs GLM: benchmarks and API costs

GLM-5.3 leads SciCode at 59.03% versus 50.58%. Qwen3.8-Flash-Next has the lower illustrated request cost at $0.0025; the other model costs $0.0184.

Updated · 6 sources · Methodology

DimensionWinnerMarginEvidence
SciCode (%)GLM-5.38.45 ppQwen3.8-Flash-Next 50.58% / GLM-5.3 59.03%12
Output speed (tokens/s)GLM-5.318.8 tokens/sQwen3.8-Flash-Next 55.4 / GLM-5.3 74.2 tokens/s
AA time to first answer token (s)GLM-5.39.01 sQwen3.8-Flash-Next 38.17 s / GLM-5.3 29.16 s
Estimated request costQwen3.8-Flash-Next$0.015910K + 1K: Qwen3.8-Flash-Next $0.0025 / GLM-5.3 $0.018443

AA coding quality benchmarks

MetricQwen3.8-Flash-NextGLM-5.3 (Max)
SciCode (%)50.58159.032

Performance

MetricQwen3.8-Flash-NextGLM-5.3 (Max)
Output speed (tokens/s)55.474.2
AA time to first answer token (s)38.1729.16

Prices and request cost

MetricQwen3.8-Flash-Next · LyceumGLM-5.3 · Novita AI
Input / 1M tokens$0.23$1.44
Output / 1M tokens$0.53$4.44
Context window (tokens)262,14451,048,5764
Max output (tokens)131,0726131,0724
API documentationNovita AI · Chat API
Cost example: 10,000 input + 1,000 billed output tokens
Qwen3.8-Flash-Next$0.0025
GLM-5.3$0.0184

Coding, reasoning, context and cost

Coding benchmark

  • AA SciCode: Qwen3.8-Flash-Next 50.58% / GLM-5.3 59.03%.12

Reasoning benchmark

  • The configured AA first-answer wait is 38.17 s for Qwen3.8-Flash-Next and 29.16 s for GLM-5.3.12

Context window (tokens)

  • GLM accommodates 1,048,576 context tokens rather than Qwen’s 262,144, allowing larger inputs on the listed endpoint.12

Estimated request cost

  • The budget tradeoff is $0.0184 for GLM versus $0.0025 for Qwen; choose Qwen when this token scenario sets the spending limit.43

Methodology and sources

SciCode is AA’s scientific-coding evaluation, reported as a percentage; higher is better. It is separate from the 10,000-input-token performance test and is not an in-editor Cline evaluation. Quality and performance use the configurations shown beside the model names. A current AA v5 Intelligence Index is displayed only where both models have a published value. AA performance uses the 10,000-input-token benchmark at each model’s displayed configuration; reasoning settings vary by model. Output speed is the median output rate. Answer latency measures the first answer token after reasoning, not the initial streamed reasoning chunk. For non-streamed reasoning and non-reasoning models, AA reports the same first-chunk and first-answer event. AA estimates answer latency from initial response and reasoning duration; it is not a separately sampled median. These are reference deployments, not Novita endpoint guarantees. The uncached cost example bills 10,000 input and 1,000 output tokens at each named seller’s tariff; it is not the cost of the AA benchmark. Prices, context and output limits use the cited seller sources.

  1. Artificial Analysis · Qwen3.8-Flash-Next · quality and performance
  2. Artificial Analysis · GLM-5.3 (Max) · quality and performance
  3. Lyceum · Qwen3.8-Flash-Next · API pricing and limits
  4. Novita AI · GLM-5.3 · endpoint pricing and limits
  5. Lyceum · Qwen3.8-Flash-Next · native context specification
  6. Lyceum · serverless request limits

Build with Novita Model APIs

Create an API key and connect your application.

Start BuildingExplore Models