GLM-4.7 Flash
Newer version available: GLM-5.3-Flash
Released
2025-01-01
Last refreshed
2026-10-27
Status
Researched 18d ago
Open sourceCommercial use: permittedCodingRAGAgentsLong contextClassificationJSON / Tool use
Zhipu AI releases · 15 in the last 12 months · this family litChangelog →
Created by
Pricing
Output / 1M
Free
Input / 1M
Free
Cheapest of 6 routes · Z.ai · cache read Free
Created by
Pricing
Output / 1M
Free
Input / 1M
Free
Cheapest of 6 routes · Z.ai · cache read Free
About
GLM-4.7 Flash is Tsinghua Knowledge Engineering Group (THUDM)'s GLM-4 model. It offers a 198K-token context window.
Providers(6)
Compare all 6Compare API pricing across 6 tracked provider routes for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Cache | Route |
|---|---|---|---|---|
| Z.ai | Free | Free | read Free | Serverless |
| Cloudflare Workers AI | $0.060 | $0.400 | - | Serverless |
| Novita AI | $0.070 | $0.400 | - | Serverless |
| OpenRouter | $0.070 | $0.400 | read $0.010 | Serverless |
| Vercel AI Gateway | $0.070 | $0.400 | - | Serverless |
| Fireworks AI | $0.900 | $0.900 | - | Serverless |
Available via routers & gateways(1)
Capabilities
ReasoningJSON / Tool useStructured Outputs
Benchmark peer barsfor Coding
SWE-bench VerifiedRank 89 of 104
LiveCodeBenchRank 67 of 79
Benchmark scores(8)
Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| 91.6 | From official Z.ai HuggingFace model cardObserved 2026-10-06 | — | ||
| 60.5 | diamondObserved 2026-08-30 | Configuration: glm-4.7-flash (Epoch row without effort suffix; reasoning setting not labeled) Harness: Epoch AI Inspect harness (GPQA Diamond, 198 Q) Evaluator: Epoch AI Confidence: confirmed Notes: Epoch AI-run evaluation, from the Epoch AI Benchmarking Hub data download https://epoch.ai/data/benchmark_data.zip (file gpqa_diamond.csv, row 'glm-4.7-flash', run started 2026-08-30). Exact mean score 0.6054292929 (stderr 0.0293); stored value is rounded to 2 decimals; other Epoch configurations of this model in the same file: glm-4.7-flash_none 0.4514. Epoch data is CC BY 4.0: credit ‘Epoch AI, Capabilities & benchmarking, https://epoch.ai/benchmarks’. Vendor claim: Z.ai reports 75.2 in the GLM-4.7-Flash Hugging Face model card (variant not stated). | ||
| 64.0 | v6Observed 2026-10-06 | — | ||
| 14.4 | From official Z.ai HuggingFace model card (tools setting not stated)Observed 2026-10-06 | — | ||
| 59.2 | From official Z.ai HuggingFace model cardObserved 2026-10-06 | — | ||
| 42.8 | From official Z.ai HuggingFace model cardObserved 2026-10-06 | — | ||
| 1365.0 | text_style_control overall 2026-10-22 (style control, default view)Observed 2026-10-26 | Configuration: GLM-4.7-Flash Harness: LMArena blind pairwise human votes, Bradley-Terry Evaluator: LMArena Confidence: confirmed Notes: LMArena text style control leaderboard, overall category, published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GLM-4.7-Flash': rank 203, Arena Score 1364.7 (95% CI 1359.0-1370.4), 12013 votes. Retrieved 2026-10-26. | Source Source: LMArena (lmarena.ai), CC BY 4.0 | |
| 1296.0 | text creative_writing 2026-10-22Observed 2026-10-22 | Configuration: GLM-4.7-Flash Harness: LMArena blind pairwise human votes, Bradley-Terry Evaluator: LMArena Confidence: confirmed Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'GLM-4.7-Flash' (category creative_writing in text/latest-00000-of-00001.parquet): rank 223 of 416, rating 1296.0 (95% CI 1282.3-1309.6), 1873 votes. Source: LMArena (CC BY 4.0). Retrieved 2026-10-27. | Source Source: LMArena (lmarena.ai), CC BY 4.0 |
Migration checks
No linked migration route is available for this model yet.