[mle]/DaBench
Leaderboard

Qwen3.8 27B

qwen/qwen3.8-27b

Hostedqwen196 tok/sopenrouterserved by Near AIFP8256K context
Run recordCompareOpen weights
Overall score
48

79 of 153 test cases fully passed

Watch the test

Qwen3.8 27B Local Test | New Open Weight 16GB Model King? | Coding with OpenCode and Agentic Work

Watch on YouTube

Scores by task

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

  1. 57Agents11 of 21 passedp90 latency 59.6s
  2. 73Structured Extraction18 of 21 passedp90 latency 38.1s
  3. 37Coding5 of 21 passedp90 latency 1m 18s
  4. 61Grounded Answers14 of 21 passedp90 latency 1m 17s
  5. 67Instruction Following11 of 21 passedp90 latency 58.0s
  6. 36Business Logic10 of 21 passedp90 latency 53.7s
  7. 51Security & permissions10 of 21 passedp90 latency 46.9s
  8. 0Diagnosis & investigation0 of 6 passedp90 latency 1m 16s

Run it yourself

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

Deployment
Hosted
API
OpenRouter
Serving provider
Near AI
Precision
FP8
Context window
256K
Throughput
196 tok/s
Cost of a suite run
$1.28
Cost per case
$0.0083
Time per case
33.3s
Measured
2026-10-12
Run this buildOpenRouterHosted
curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen/qwen3.8-27b","provider":{"order":["near-ai/fp8"],"allow_fallbacks":false},"service_tier":"default","messages":[{"role":"user","content":"Hello"}]}'

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

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