Side-by-side comparison · models
GPT OSS 120B vs GLM-4.7-Flash
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
GPT OSS 120B
Open GPT reasoning model for self-hosted agents and controllable deployments.
Models
GLM-4.7-Flash
Efficient GLM model for fast reasoning, coding, and agent workflows.
Checked facts
Line by line
| Measure | GPT OSS 120B | GLM-4.7-Flash |
|---|---|---|
| Input price | $0.150 per million tokens | $0.060 per million tokens |
| Output price | $0.600 per million tokens | $0.400 per million tokens |
| Price on akashml | $0.030 per million cached input tokens | Not checked yet |
| Price on akashml | $0.030 per million input tokens | Not checked yet |
| Price on akashml | $0.187 per million output tokens | Not checked yet |
| Price on amazon-bedrock | $0.150 per million input tokens | Not checked yet |
| Price on amazon-bedrock | $0.600 per million output tokens | Not checked yet |
| Price on baseten | $0.100 per million cached input tokens | Not checked yet |
| Price on baseten | $0.100 per million input tokens | Not checked yet |
| Price on baseten | $0.500 per million output tokens | Not checked yet |
| Price on cerebras | $0.350 per million cached input tokens | Not checked yet |
| Price on cerebras | $0.350 per million input tokens | Not checked yet |
| Price on cerebras | $0.750 per million output tokens | Not checked yet |
| Price on coreweave | $0.030 per million cached input tokens | Not checked yet |
| Price on coreweave | $0.030 per million input tokens | Not checked yet |
| Price on coreweave | $0.170 per million output tokens | Not checked yet |
| Price on crusoe | $0.050 per million cached input tokens | Not checked yet |
| Price on crusoe | $0.050 per million input tokens | Not checked yet |
| Price on crusoe | $0.250 per million output tokens | Not checked yet |
| Price on deepinfra | $0.037 per million input tokens | Not checked yet |
| Price on deepinfra | $0.170 per million output tokens | Not checked yet |
| Price on dekallm | $0.030 per million cached input tokens | Not checked yet |
| Price on dekallm | $0.030 per million input tokens | Not checked yet |
| Price on dekallm | $0.180 per million output tokens | Not checked yet |
| Price on digitalocean | $0.012 per million cached input tokens | Not checked yet |
| Price on digitalocean | $0.060 per million input tokens | Not checked yet |
| Price on digitalocean | $0.420 per million output tokens | Not checked yet |
| Price on google-vertex | $0.090 per million input tokens | Not checked yet |
| Price on google-vertex | $0.360 per million output tokens | Not checked yet |
| Price on groq | $0.075 per million cached input tokens | Not checked yet |
| Price on groq | $0.150 per million input tokens | Not checked yet |
| Price on groq | $0.600 per million output tokens | Not checked yet |
| Price on io-net | $0.015 per million cached input tokens | Not checked yet |
| Price on io-net | $0.030 per million input tokens | Not checked yet |
| Price on io-net | $0.150 per million output tokens | Not checked yet |
| Price on mancer | $0.055 per million cached input tokens | Not checked yet |
| Price on mancer | $0.055 per million input tokens | Not checked yet |
| Price on mancer | $0.250 per million output tokens | Not checked yet |
| Price on mara | $0.150 per million input tokens | Not checked yet |
| Price on mara | $0.750 per million output tokens | Not checked yet |
| Price on near-ai | $0.015 per million cached input tokens | Not checked yet |
| Price on near-ai | $0.030 per million input tokens | Not checked yet |
| Price on near-ai | $0.150 per million output tokens | Not checked yet |
| Price on nebius | $0.100 per million cached input tokens | Not checked yet |
| Price on nebius | $0.150 per million input tokens | Not checked yet |
| Price on nebius | $0.600 per million output tokens | Not checked yet |
| Price on novita | $0.050 per million input tokens | $0.070 per million input tokens |
| Price on novita | $0.250 per million output tokens | $0.400 per million output tokens |
| Price on parasail | $0.055 per million cached input tokens | Not checked yet |
| Price on parasail | $0.100 per million input tokens | Not checked yet |
| Price on parasail | $0.750 per million output tokens | Not checked yet |
| Price on phala | $0.150 per million input tokens | Not checked yet |
| Price on cloudflare | Not checked yet | $0.060 per million input tokens |
| Price on cloudflare | Not checked yet | $0.400 per million output tokens |
| Price on novita | Not checked yet | $0.010 per million cached input tokens |
| Latest version | openai/gpt-oss-120b | z-ai/glm-4.7-flash-20260119 |
| Time to first token (ms) | 810 | 1,020 |
| Time to first token (ms) on akashml | 548 | Not checked yet |
| Time to first token (ms) on amazon-bedrock | 517 | Not checked yet |
| Time to first token (ms) on baseten | 180 | Not checked yet |
| Time to first token (ms) on cerebras | 228 | Not checked yet |
| Time to first token (ms) on coreweave | 277 | Not checked yet |
| Time to first token (ms) on crusoe | 331 | Not checked yet |
| Time to first token (ms) on deepinfra | 501 | Not checked yet |
| Time to first token (ms) on dekallm | 501 | Not checked yet |
| Time to first token (ms) on digitalocean | 570 | Not checked yet |
| Time to first token (ms) on google-vertex | 2,313 | Not checked yet |
| Time to first token (ms) on groq | 240 | Not checked yet |
| Time to first token (ms) on io-net | 847 | Not checked yet |
| Time to first token (ms) on mancer | 595 | Not checked yet |
| Time to first token (ms) on mara | 1,159 | Not checked yet |
| Time to first token (ms) on near-ai | 599 | Not checked yet |
| Time to first token (ms) on nebius | 228 | Not checked yet |
| Time to first token (ms) on novita | 832 | 1,207 |
| Time to first token (ms) on parasail | 472 | Not checked yet |
| Time to first token (ms) on phala | 969 | Not checked yet |
| Time to first token (ms) on prime-intellect | 772 | Not checked yet |
| Time to first token (ms) on sail-research | 1,064 | Not checked yet |
| Time to first token (ms) on sambanova | 816 | Not checked yet |
| Time to first token (ms) on siliconflow | 1,035 | Not checked yet |
| Time to first token (ms) on together | 183 | Not checked yet |
| Time to first token (ms) on venice | 799 | Not checked yet |
| Output speed (tokens a second) | 210.3 | 122.5 |
| Output speed (tokens a second) on akashml | 59 | Not checked yet |
| Output speed (tokens a second) on amazon-bedrock | 202 | Not checked yet |
| Output speed (tokens a second) on baseten | 140 | Not checked yet |
| Output speed (tokens a second) on cerebras | 604 | Not checked yet |
| Output speed (tokens a second) on coreweave | 82 | Not checked yet |
| Output speed (tokens a second) on crusoe | 149 | Not checked yet |
| Output speed (tokens a second) on deepinfra | 27 | Not checked yet |
| Output speed (tokens a second) on dekallm | 48 | Not checked yet |
| Output speed (tokens a second) on digitalocean | 30 | Not checked yet |
| Output speed (tokens a second) on google-vertex | 50 | Not checked yet |
| Output speed (tokens a second) on groq | 301 | Not checked yet |
| Output speed (tokens a second) on io-net | 40 | Not checked yet |
| Output speed (tokens a second) on mancer | 106.5 | Not checked yet |
| Time to first token (ms) on cloudflare | Not checked yet | 315 |
| Output speed (tokens a second) on cloudflare | Not checked yet | 44 |
| Output speed (tokens a second) on novita | Not checked yet | 54 |
| Tokens served on OpenRouter in a day | 69B (for 10 Nov 2026) | 1.3B (for 10 Nov 2026) |
| Tool calls on OpenRouter in a day | 761.8K (for 10 Nov 2026) | 51.9K (for 10 Nov 2026) |
| Artificial Analysis Intelligence Index | 3.8 | 14.9 |
| Artificial Analysis Coding Index | 30.4 | Not checked yet |
| Artificial Analysis Agentic Index | 3.7 | Not checked yet |
| LMArena text rating (Elo) | 1,365 | 1,351 |
| Design Arena rating (Elo) | 986 | 1,202 |
| Gpqa diamond | 75.8% (for 5 Aug 2025) | Not checked yet |
| Terminal bench | 18.7% (for 5 Aug 2025) | Not checked yet |
| Aider polyglot | 41.8% (for 5 Aug 2025) | Not checked yet |
| Apex agents | 4.4% (for 5 Aug 2025) | Not checked yet |
| Dtbench | 76.3% (for 5 Aug 2025) | Not checked yet |
| Fiction livebench | 44.4% (for 5 Aug 2025) | Not checked yet |
| Lech mazur writing | 77.1% (for 5 Aug 2025) | Not checked yet |
| Lmca | 22.1% (for 5 Aug 2025) | Not checked yet |
| Metr time horizons | 56.6% (for 5 Aug 2025) | Not checked yet |
| Mystery game puzzles | 2% (for 5 Aug 2025) | Not checked yet |
| Simplebench | 22.1% (for 5 Aug 2025) | Not checked yet |
| Surface evolver bench | 25% (for 5 Aug 2025) | Not checked yet |
| Weirdml | 48.2% (for 5 Aug 2025) | Not checked yet |
| Otis mock aime 2024 2025 | 88.9% (saturated benchmark: the top models all score near the ceiling) (for 5 Aug 2025) | Not checked yet |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.