Z.ai, released Sep 30, 2025
GLM-4.6price, context, benchmarks and release details
Provisional: not enough results to rank yet Open weights 55% confidence 55 percent, Medium confidence, 4 of 9 expected sources in
47.8
SI Score
#114 of 142 ranked models
- Input, per 1M tokens
- $0.60Z.ai API pricingOfficial Z.ai global Standard uncached token rate; coding-plan subscriptions, cache and Batch excluded
Retrieved Oct 9, 2026 · factual citation
Open source ↗ - Output, per 1M tokens
- $2.20Z.ai API pricingOfficial Z.ai global Standard uncached token rate; coding-plan subscriptions, cache and Batch excluded
Retrieved Oct 9, 2026 · factual citation
Open source ↗ - Context window
- 205Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Max output
- 131Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Released
- Sep 30, 2025models.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗
How the score breaks down
Coding (weight 40 percent) 33.0
Math (weight 15 percent) —
Preference (weight 15 percent) 76.8
Reasoning (weight 30 percent) —
Weights: reasoning 30%, math 15%, coding 40%, preference 15%. Results use fixed 0–100 scales before averaging, and thin evidence is pulled toward 50. Method si-v3-retained-evidence-2, computed Oct 9, 2026, 05:13 UTC.
Around it on the leaderboard
- 112 QwQ 32B 48.2
- 113 DeepSeek-R1 47.9
- 114 GLM-4.6 47.8
- 115 Claude Sonnet 4 46.6
- 116 Nemotron 3.5 Lightning 30B A3B 46.1
Benchmark results
4 benchmarks, 6 resultsEach row shows the best published result. Where a model was tested at several settings, such as reasoning effort, open the row to see each one. Hover or tap a value for its source.
SWE-bench Pro (public)coding 9.7%Official model cards via models.devLab-reported; metric resolve rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] public
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗
9.7 2 settings
- Setting 1 9.7%Official model cards via models.devLab-reported; metric resolve rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] public
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ - Setting 2 9.7%SWE-bench Pro (public)Published steward score [variant] Published Jan 27, 2026
Retrieved Oct 9, 2026 · factual citation
Open source ↗
SWE-bench Verifiedcoding 68.2%SWE-bench VerifiedSWE-bench published model plus agent result; harness retained, not a base model evaluation [variant] UndisclosedPublished Sep 30, 2025
Retrieved Oct 9, 2026 · factual citation
Open source ↗
68.2 2 settings
- Setting 1 55.4%SWE-bench VerifiedSWE-bench published model plus agent result; harness retained, not a base model evaluation [variant] mini-SWE-agent; 1.17.1Published Dec 1, 2025
Retrieved Oct 9, 2026 · factual citation
Open source ↗ - Setting 2 68.2%SWE-bench VerifiedSWE-bench published model plus agent result; harness retained, not a base model evaluation [variant] UndisclosedPublished Sep 30, 2025
Retrieved Oct 9, 2026 · factual citation
Open source ↗
Terminal-Benchcoding 25%Official model cards via models.devLab-reported; metric success rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] Published May 22, 2026
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ 25.0
Open source ↗ 25.0
LMArena Textpreference 1439.8 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 8, 2026
Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗ 76.8
Open source ↗ 76.8
Normalization uses a fixed 0–100 scale for each unit, independent of other models. Compare evaluation conditions before reading a small gap as decisive. “Lab-reported” marks the provider's own published figure.
Details and sources
- Open weights
- Yesmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - License
- not yet reported
- Input modalities
- textmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - First seen by SuperIndex
- Oct 8, 2026
- Coverage
- 44% of expected source weight
Reported (4)
- Official model cards via models.devOct 8, 2026
- LMArena / ArenaOct 8, 2026
- SWE-bench VerifiedOct 8, 2026
- SWE-bench Pro (public)Oct 8, 2026
Awaiting (5)
- ARC Prize10% of weight
- Epoch AI Benchmarking20% of weight
- Humanity’s Last Exam10% of weight
- LiveBench10% of weight
- Terminal-Bench5% of weight
Confidence rises as pending sources publish. Some sources never cover some models, so confidence reaches 100% at 80% of expected weight.