Benchmark avg
48.6
17 tracked scores
Model portfolio
Tap a star to rate
Microsoft · Open · Context 128K tokens
Language Cap
119
Cadle Language Cap 119 · 45%
Benchmark average
—
Not enough benchmark results yet
Pricing
Unlisted
no list price
Only measured signals
Benchmark avg
48.6
17 tracked scores
Strongest signal
arc-ai2
Score 88.8
Blended $/1M
Unlisted
No published list price
Context
128K
Provider context window (not max output)
PROVIDER
Microsoft
CONTEXT
128K tokens
STATUS
ACTIVE
Real operator evaluations, prompt robustness observations, and deployment notes for Phi-3-medium-128k-instruct.
Tested Phi-3-medium-128k-instruct on high-concurrency tool loops. Token generation remains consistent under sustained context load, with minimal prompt drift on complex JSON schemas.
Evaluated Phi-3-medium-128k-instruct on multi-turn code synthesis. Code reasoning and diff application succeed on ~88% of test suites. Best paired with strict system prompt guardrails.
| Metric | |
|---|---|
| Provider | Microsoft |
| Release | |
| Input / 1M | No data yet |
| Output / 1M | No data yet |
| Context | 128K tokens |
| Cadle Cap | 119 |
| Cap board | Cadle Language Cap 119 |
| bigcodebench | 48.7 |
| mmlu-pro | 41.2 |
| musr | 11.4 |
| gpqa-diamond | 11.5 |
| math-level-5 | 19.2 |
| bbh | 48.5 |
| ifeval | 60.4 |
| winogrande | 63.0 |
| openbookqa | 83.2 |
| mmlu | 70.7 |
| anli | 33.7 |
| hellaswag | 76.5 |
Scores may use different scales — check the Benchmarks tab for source links.
Evidence linked to this model — listing, hub, benchmarks, or live gateway prices.
API
Microsoft listing