[mle]/DaBench
Leaderboard

GLM 5.3 Flash

z-ai/glm-5.3-flash

Hostedz-ai60 tok/sopenrouterserved by StreamLakeFP81M context
Run recordCompare
Overall score
35

48 of 153 test cases fully passed

Scores by task

Scores give partial credit. Passed counts fully correct answers only. Select a task to read its test cases.

  1. 10Agents2 of 21 passedp90 latency 1m 1s
  2. 48Structured Extraction10 of 21 passedp90 latency 2m 53s
  3. 54Coding6 of 21 passedp90 latency 4m 43s
  4. 37Grounded Answers8 of 21 passedp90 latency 3m 41s
  5. 43Instruction Following6 of 21 passedp90 latency 1m 7s
  6. 33Business Logic7 of 21 passedp90 latency 3m 0s
  7. 55Security & permissions9 of 21 passedp90 latency 56.2s
  8. 0Diagnosis & investigation0 of 6 passedp90 latency 36.6s

Run it yourself

The API call the benchmark made, and the settings behind this score.

Deployment
Hosted
API
OpenRouter
Serving provider
StreamLake
Precision
FP8
Context window
1M
Throughput
60 tok/s
Cost of a suite run
$0.19
Cost per case
$0.0013
Time per case
55.5s
Measured
2026-10-03
Run this buildOpenRouterHosted
curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"z-ai/glm-5.3-flash","provider":{"order":["streamlake/fp8"],"allow_fallbacks":false},"messages":[{"role":"user","content":"Hello"}]}'

The request the benchmark sent, pinned to the provider that served it.

Watch the test

Video tests of this model on coding and agent work.

GLM-5.3-Flash First Test | Better Than Qwen3.8 and DeepSeek v4? | Coding Tests with OpenCode29 Aug 2026

One email when a new model lands on the board or in the showcase. Unsubscribe anytime.

LeaderboardCompare GLM 5.3 Flash with another model