{ "as_of": "2026-10-01", "use_for_chart_calibration": true, "reports": [ { "provider": "zai", "author": "dontWORRYimASIAN", "source": "https://www.reddit.com/r/ZaiGLM/comments/1w4eg4b/comment/p773rtu/", "report_date": null, "accessed_as_of": "2026-10-01", "evidence_type": "First-hand self-report; no independently audited invoice or telemetry", "plan": "GLM Coding Plan Lite", "monthly_fee_usd_reported": 18, "five_hour_credits_reported": 2000, "weekly_credits_reported": 10000, "sample_duration_days_approx": 3, "tokens_reported_range": [ 30000000, 45000000 ], "calls_reported_range": [ 500, 600 ], "credits_used_reported_range": [ 3400, 5700 ], "weekly_utilization_percent_reported_range": [ 35, 57 ], "weekly_utilization_percent_computed_from_credits": [ 34, 57 ], "offpeak_usage_fraction_reported": 0.95, "model_mix_reported": { "GLM-5.3": 0.43, "GLM-5.3-Flash": 0.57 }, "harness": "Hermes; coding within OMP within Hermes", "api_equivalent_monthly_usd_reported_actual_august": [ 35, 55 ], "api_equivalent_monthly_usd_reported_full_utilization_projection": [ 95, 165 ], "cash_cost_for_sample": null, "monthly_plan_utilization_measured": null, "apply_to_chart_capacity": false, "limitations": [ "Mixed models and different harness from the AA GLM-5.3 OpenCode configuration", "Token/call/credit figures are broad ranges and sample versus month periods differ", "Reported cache percentages use inconsistent or unspecified denominators", "Author mentions a historical Flash API discount; not a current provider API quota", "Reported weekly share is rounded; supplied credit range calculates to 34\u201357%, not exactly 35\u201357%", "Full-use monthly API-equivalent range is the author\u2019s projection, not a measured allowance" ] }, { "provider": "openai", "author": "IndividualPlus2011", "source": "https://www.reddit.com/r/codex/comments/1wtlkg6/gpt_61_brought_another_nerf_to_usage_table_update/", "method_source": "https://www.reddit.com/r/codex/comments/1wp0v6x/allowance_for_gpt_6_luna_sol_and_astra_measured/", "report_date": "2026-09-29", "report_date_basis": "Explicit 6.1 edit date in method post; update page relative publication age differs", "plan": "ChatGPT Plus", "monthly_fee_usd_reported": 20, "model": "GPT-6.1 Sol", "harness": "Pi", "api_equivalent_usd_per_fresh_five_hour_window": 7.5, "five_hour_window_utilization_reported": 1, "weekly_api_equivalent_usd_reported_projection": 45, "monthly_api_equivalent_usd_reported_30_day_projection": 193, "monthly_api_equivalent_usd_52_week_projection": 195.0, "cash_cost_for_sample": null, "apply_to_chart_capacity": false, "limitations": [ "Only Plus was tested; Pro scaling in the post is not measured", "Weekly and monthly totals are extrapolations from a five-hour-window observation", "No long context; workload/cache mix and effort can differ from the benchmark", "Current new-buyer Pro 200 allowances differ from grandfathered historical accounts" ] }, { "provider": "anthropic", "source": "https://www.remakebench.com/capacity/plans/claude-max-20x", "report_date": "2026-09-25", "plan": "Owner-reported Claude Max 20x; app label Pro conflicts", "monthly_fee_usd_reported": 200, "model_mix_opus_55_fraction_reported": 0.963, "sample_window_start": "2026-09-24T00:59:59Z", "sample_window_reset": "2026-10-01T00:59:59Z", "api_equivalent_usd_corrected_sample": 1249.83, "weekly_utilization_percent_reported": 58, "weekly_api_equivalent_usd_inferred": 2154.8793103448274, "monthly_api_equivalent_usd_52_week_projection": 9337.810344827585, "cash_cost_for_sample": null, "apply_to_chart_capacity": false, "limitations": [ "Owner fee/Max tier conflicts with app Pro label", "One rounded weekly meter reading; model-mix dependence", "Corrected receipts still may omit activity outside local transcripts", "Extra usage had been charged; historical totals are not clean included-limit ceilings", "Harness cohort differs from Artificial Analysis; no score transferred" ] }, { "provider": "xai", "source": "https://www.remakebench.com/capacity/plans/supergrok-heavy", "report_date": "2026-09-25", "plan": "SuperGrok Heavy", "monthly_fee_usd_reported": 300, "model": "Grok 4.7", "effort_mix_xhigh_fraction_reported": 0.961, "harness": "Grok CLI", "meter_start_percent_used": 56, "meter_end_percent_used": 75, "meter_delta_percentage_points": 19, "sample_start": "2026-09-23T18:22:15Z", "sample_end": "2026-09-25T09:09:24Z", "api_equivalent_usd_deduplicated_sample": 76.09728984, "weekly_api_equivalent_usd_inferred": 400.5120517894737, "monthly_api_equivalent_usd_52_week_projection": 1735.5522244210526, "cash_cost_for_sample": null, "apply_to_chart_capacity": false, "limitations": [ "Weekly pool shared across Grok products; unlogged use may affect meter delta", "Mostly XHigh rather than one pure configuration", "API-equivalent valuation is not cash billed or provider-guaranteed dollars", "Model/effort and workload differ from the AA Grok Build benchmark configuration" ] }, { "provider": "anthropic", "author": "oxmannnn", "source": "https://www.reddit.com/r/ClaudeAI/comments/1wtagdd/sonnet_55_did_this_opus_55_quality_with_half_price/", "quota_comment_source": "https://www.reddit.com/r/ClaudeAI/comments/1wtagdd/comment/pcsnpqx/", "report_date": null, "plan": "Claude Max 20x, self-reported", "monthly_fee_usd_reported": null, "model": "Sonnet 5.5", "api_equivalent_usd_30_second_video": 30.46, "api_equivalent_usd_both_video_versions": 35.4, "sample_duration_hours_reported": 2.4, "tool_calls_reported": 779, "weekly_utilization_percent_measured": null, "quota_comment_not_independently_accessible": false, "cash_cost_for_sample": null, "apply_to_chart_capacity": true, "limitations": [ "One self-reported rounded ~2% weekly share; no raw before/after quota-meter logs", "Unclear whether quota share covers the $30.46 final video or $35.40 both-version total", "Creator Ultracode/subagent workload differs from AA Claude Code max benchmark configuration", "Transcript-based API valuation is not a billed amount or guaranteed quota", "Selected 1\u20133% / $30.46\u2013$35.40 sensitivity envelope is not a confidence interval; structural uncertainty remains unquantified", "Same-token counterfactual Opus valuation is not an observed Opus transcript ledger" ], "weekly_utilization_percent_reported_approximate": 2, "monthly_fee_usd_official": 200, "fee_source": "https://support.anthropic.com/en/articles/11049741-what-is-the-max-plan", "harness": "Claude Desktop / Claude Code, self-reported Ultracode and 14 subagents", "calibration_status": "Provisional selected scenario only; rounded self-report with workload-scope and harness/effort transfer uncertainty", "corroboration_without_priced_ledger": [ { "source": "https://www.reddit.com/r/ClaudeCode/comments/1wt2vw6/sonnet_55_is_good/?sort=new", "author": "Dingenjor", "plan": "Max 20x, self-reported", "model": "Sonnet 5.5", "effort": "medium", "hours_reported": 6, "weekly_meter_start_percent_reported": 92, "weekly_meter_end_percent_reported": 93, "api_equivalent_usd": null, "apply_to_chart_capacity": false }, { "source": "https://www.reddit.com/r/ClaudeCode/comments/1wt2vw6/sonnet_55_is_good/?sort=new", "author": "thehoundtrainer", "plan": "Max 20x, self-reported", "model": "Sonnet 5.5", "effort": "max", "hours_reported": 8, "weekly_share_percent_reported": 15, "api_equivalent_usd": null, "apply_to_chart_capacity": false } ], "crosspost_deduplication": { "author": "oxmannnn", "independent_priced_quota_samples": 1, "crossposts": [ "https://www.reddit.com/r/ClaudeAI/comments/1wtagdd/sonnet_55_did_this_opus_55_quality_with_half_price/", "https://www.reddit.com/r/ClaudeCode/comments/1wtapu5/sonnet_55_built_a_30_s_motiongraphics_reel_same/", "https://www.reddit.com/r/vibecoding/comments/1wtajpx/sonnet_55_is_the_same_quality_as_opus_55_but/" ], "note": "Same video/run/token ledger across three communities. Third post associates Max20x, about $35 whole-session API equivalent and around 2% weekly meter explicitly. This improves scope interpretation but is not a replicate." }, "normalization_scenarios": { "api_dollar_transfer": [ 0.3700339190709166, 0.43004598605090105 ], "token_weight_transfer": [ 0.5495389925971031, 0.6803590380199507 ] }, "selected_band": [ 0.1850169595354583, 1.020538557029926 ], "alternate_normalization_source": "https://alldonesites.com/data/claude-usage.json" } ], "glm_schedule_sensitivities": { "fee_usd": 80, "monthly_credits": 260000, "offpeak_main_per_attempt": 0.41444475858170243, "offpeak_fraction_95_percent_per_attempt": 0.43516699651078755, "uniform_168_hour_week_per_attempt": 0.46378342031761943, "all_peak_per_attempt": 0.8288895171634049, "method": "Weighted credit multiplier holding AA token mix and full weekly utilization fixed; percentages are assumptions, not observed benchmark timing" }, "calibration_scope": "Only the Sonnet Max 20x report supplies an explicitly provisional scenario; other four reports remain outside chart capacity calibration." }